Combinations of short genetic variants in the diagnosis of alzheimer disease

By employing specific combinations of short genetic variants and advanced analytical methods, the diagnostic accuracy for Alzheimer's disease is improved, overcoming the limitations of traditional genetic risk score approaches by capturing synergistic effects and enhancing predictive capabilities.

WO2024231499A1PCT designated stage expired Publication Date: 2024-11-14VLAAMS INTERUNIVERSITAIR INST VOOR BIOTECHNOLOGIE VZW +1
View PDF 10 Cites 0 Cited by

Patent Information

Application Number
PCT/EP2024/062838
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-10
Filing Date
2024-05-08
Publication Date
2024-11-14

AI Technical Summary

Technical Problem

Current diagnostic methods for Alzheimer's disease (AD) are inadequate in distinguishing between AD patients and healthy subjects due to the additive or linear effects of single nucleotide polymorphisms (SNPs) in genetic risk scores, which fail to capture the synergistic effects of short genetic variants (SGVs) and thus cannot accurately predict or diagnose AD risk.

Method used

The use of specific combinations of short genetic variants (SGVs) identified through multifactor dimensionality reduction (MDR) and logistic regression analysis to diagnose AD, where the presence or number of pathogenic alleles in these combinations is determined to diagnose AD or predict AD risk, incorporating synergistic effects beyond simple additive models.

Benefits of technology

This approach allows for accurate diagnosis and prediction of AD risk by identifying synergistic SGV combinations that were not apparent through traditional additive models, enhancing the ability to distinguish AD patients from healthy subjects.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure IMGF000075_0001
    Figure IMGF000075_0001
  • Figure IMGF000076_0001
    Figure IMGF000076_0001
  • Figure IMGF000077_0001
    Figure IMGF000077_0001
Patent Text Reader

Abstract

The invention relates to combinations of Alzheimer disease risk loci, more in particular short genetic variant loci (SGVs). Combinations of such SGVs were identified that are significantly and synergistically pathogenic. Applications of these combinations, such as diagnostic applications, are included as well.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] COMBINATIONS OF SHORT GENETIC VARIANTS IN THE DIAGNOSIS OF

[0002] ALZHEIMER DISEASE

[0003] FIELD OF THE INVENTION

[0004] The invention relates to combinations of Alzheimer disease risk loci, more in particular short genetic variant loci (SGVs). Combinations of such SGVs were identified that are significantly and synergistically pathogenic. Applications of these combinations, such as diagnostic applications, are included as well.

[0005] BACKGROUND OF THE INVENTION

[0006] The heritability Alzheimer disease (AD) is estimated to be between 60% and 80%. Thus, beside age, genetic risk factors are thought to be important. Although high-throughput genomic techniques resulted in identification of a large number of putative AD-associated loci / genes, much of the underlying heritability remains unexplained. In a recent genome-wide association study (GWAS) a new genetic risk score associated with the risk of future AD / dementia or progression from mild cognitive impairment to AD / dementia was developed; with this risk score based on the withheld risk alleles / SNPs a 1.6- to 1.9- fold increase in AD risk was predicted (Bellenguez et al. 2022, Nature Genetics 54:412-436). Genetic risk score (GRS, also known as polygenic risk score) analysis can be applied but such analysis segregate subjects with a different genetic background, therewith masking or excluding identification of predictive or diagnostic combinations of pathogenic alleles which on their own are only effectuating a low AD risk. By simply looking at the number of risk alleles present in a subject, it is not possible to distinguish a subject having AD or being at risk of having AD from a healthy subject. The genetic risk score being based on additive or linear effects of the identified SNPs is not sufficiently powerful to distinguish between AD and non-AD.

[0007] SUMMARY OF THE INVENTION

[0008] The present disclosure relates to methods of diagnosing a subject as Alzheimer disease (AD) patient or as potential AD patient, or as subject being at risk of developing AD, such methods including the steps of: determining the presence or absence of a pathogenic allele of the individual short genetic variants (SGVs) of a SGV combination in a DNA sample obtained from a subject, wherein the SGV combination is chosen from the SGV combinations as listed in Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table IDA, or Table 11A; diagnosing a subject as AD patient or as potential AD patient, or as subject being at risk of developing AD, when a pathogenic allele is present in each of the individual SGVs of the SGV combination. In one embodiment thereto, in such methods the presence or absence of a pathogenic allele of the individual SGVs of more than one SGV combination is determined, and the subject is diagnosed as AD patient or as potential AD patient, or as subject being at risk of developing AD, when a pathogenic allele is present in each of the individual SGVs of each of the more than one SGV combination.

[0009] The present disclosure further relates to methods of diagnosing a subject as Alzheimer disease (AD) patient or as potential AD patient, or as subject to be at risk of developing AD, such methods including the steps of: determining the number of pathogenic alleles of the SGVs of a SGV combination in a DNA sample obtained from a subject, wherein the SGV combination is chosen from the SGV combinations as listed in Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table IDA, or Table 11A; diagnosing a subject as AD patient or as potential AD patient, or as subject being at risk of developing AD, when in each of the individual SGVs of the SGV combination the number of pathogenic alleles as listed for the SGV combination chosen from Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table IDA, or Table 11A, is present.

[0010] In one embodiment thereto, in such methods the number of pathogenic alleles of the individual SGVs of more than one SGV combination is determined, and the subject is diagnosed as AD patient or as potential AD patient, or as subject being at risk of developing AD, when in each of the individual SGVs of the more than one SGV combination the number of pathogenic alleles as listed for the more than one SGV combination chosen from Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table IDA, or Table 11A, is present.

[0011] In an embodiment to any of the above methods, the pathogenic allele of an individual SGV is substituted for a SGV in linkage disequilibrium with the pathogenic allele of the individual SGV.

[0012] The present disclosure further relates to methods of screening for AD drugs / medicaments, such methods comprising:

[0013] - culturing cells obtained from an AD patient, wherein the AD patient is selected based on a method as described hereinabove, and wherein the cultured cells are displaying an AD-related phenotype;

[0014] - contacting the cultured cells with a candidate AD drug / medicament;

[0015] - selecting as AD drug / medicament a candidate AD drug / medicament that is treating the AD- related phenotype. The present disclosure further relates to methods of selecting a subject for treatment with an AD drug / medicament, or for inclusion in a clinical trial of a candidate AD drug / medicament, wherein the subject is selected based on a method as described hereinabove.

[0016] The present disclosure further relates to computer programs having instructions which when executed cause a computing or data processing system or device to perform an above-described method, or to perform a step of above-described method.

[0017] In more detail, such computer programs are having instructions which when executed cause a computing or data processing system or device to perform a method for selecting SGV combinations for diagnosing a subject as Alzheimer disease (AD) patient or as potential AD patient, or as subject being at risk of developing AD, the method comprising one or more steps of: a first selection of SGV combinations by multifactor dimensionality reduction (MDR) analysis on a training set of subjects wherein the training set comprises AD patients and healthy subjects, and wherein the SGV combinations are selected starting from set of single SGVs that are associated with AD; logistic regression (LR) analysis of the first selected SGV combinations wherein the logistic regression incorporates as covariates the main effects of the single SGVs of the selected SGV combinations; a second selection of SGV combinations from the logistic regression analysis; validating the second selected SGV combinations in a test set of subjects wherein the test set comprises AD patients and healthy subjects; wherein the validated second selected SGV combinations are synergistic in separating an AD patient or subject at risk of developing AD from a healthy subject.

[0018] In an embodiment thereto, such computer program to perform the method is further including a first selection of SGV combinations by MDR analysis on two training sets of subjects and including LR analysis on first selected SGV combinations common to the two training sets of subjects.

[0019] In another embodiment, the computer program to perform the method is further including selection of validated SGV combinations stratified according to the number of pathogenic APOEE4 alleles present in a subject.

[0020] In yet another embodiment, the computer program to perform the method with substituting a single SGV associated with AD for a SGV in linkage disequilibrium with the single SGV associated with AD. This disclosure furthermore relates to computing or data processing systems or devices, machine readable media, or diagnostic kits comprising a means for carrying out or performing any of the abovedescribed methods, or for carrying out or performing a step of any of the above-described methods.

[0021] DETAILED DESCRIPTION

[0022] In work leading to the current disclosure, and although with an a priori unknown outcome, an approach was designed to capture biologically relevant joint effect of different risk SGVs. As such, combinations of short genetic variants (SGVs, herein also referred to in general as "SGV combination(s)" or "SGV combination(s) of interest") were selected based on their synergistic effect in / on separating / distinguishing AD patients or subjects at risk of developing AD from healthy / control subjects. Such SGV combinations are thus useful in / for diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or as potential AD patient, or as subject at risk of developing AD. The subjects at risk of developing AD more in particular are at risk because of being carrier of an SGV combination as identified herein, such SGV combination being present in / carried by an AD patient.

[0023] The selection of these SGV combinations were made either including the SGV APOE rs429358 or excluding the SGV APOE rs429358. More in particular, combinations of 2 SGVs, 3 SGVs, and of 4 SGVs were selected. More in particular, these combinations are listed / given in the Tables included hereinafter, more in particular in Tables 3B, 4B, 5B, 6B, 7B, and 8B. Alternatively, the selection of these SGV combinations were made on individuals split based on their genetic background in the APOE polymorphism, thus yielding three subgroups containing either 0, 1 or 2 risk alleles of the pathogenic isoform APOEE4 (SGV APOE rs429358). More in particular, combinations of 2 SGVs together with the specific APOE genotype were selected. More in particular, these combinations are listed / given in the Tables included hereinafter, more in particular in Tables 9A (0 risk alleles of the pathogenic isoform APOEE4), IDA (1 risk allele of the pathogenic isoform APOEE4), and 11A (2 risk alleles of the pathogenic isoform APOEE4). "Pathogenic or risk alleles" or "pathogenic or riskSGVs" or the like as referred to herein are to be understood as alleles or SGVs which have been linked with AD or with a risk of developing AD.

[0024] In one aspect of this disclosure, a computer implemented method was applied for selecting these SGV combinations of interest. In such computer implemented or computer assisted methods, at least one analysis step is performed computationally, by a computer system or via a computer program product.

[0025] Furthermore in particular, such computer implemented methods at least include an analysis step including / taking into account / incorporating the main effects of the single SGVs (part) of a SGV combination of interest as covariates. In further detail, such computer implemented methods of selecting single genetic variant (SGV) combinations for diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or as potential AD patient, or as subject at risk of developing AD, or for separating AD patients or subjects at risk of developing AD from healthy subjects, are methods comprising the steps of: identification of a set of single SGVs that separate AD patients or subjects at risk of developing AD from healthy / control subjects; a first selection of SGV combinations by multifactor dimensionality reduction (MDR) analysis on a training set of subjects wherein the training set comprises AD patients and healthy / control subjects; logistic regression (LR) analysis of the first selected SGV combinations wherein the logistic regression includes the main effect of the single SGVs as covariate; a second selection of SGV combinations from the logistic regression analysis; validating the second selected SGV combinations in a test set of subjects wherein the test set comprises AD patients and healthy subjects; wherein the validated second selected SGV combinations are synergistic in diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or as potential AD patient, or as subject at risk of developing AD, or in separating AD patients or subjects at risk of developing AD from healthy subjects. Such synergistic effect is exceeding the simple additive effect of the single SGVs in the SGV combination.

[0026] Alternatively formulated, this aspect relates to a computer program having instructions which when executed cause a computing or data processing system or device to perform a method for selecting SGV combinations for diagnosing a subject as Alzheimer disease (AD) patient or as potential AD patient, or as subject being at risk of developing AD, the method comprising the steps of: a first selection of SGV combinations by multifactor dimensionality reduction (MDR) analysis on a training set of subjects wherein the training set comprises AD patients and healthy subjects, and wherein the SGV combinations are selected starting from set of single SGVs that are associated with AD or starting from a set of single pathogenic SGVs; logistic regression (LR) analysis of the first selected SGV combinations wherein the logistic regression incorporates as covariates the main effects of the single SGVs of the selected SGV combinations; a second selection of SGV combinations from the logistic regression analysis; validating the second selected SGV combinations in a test set of subjects wherein the test set comprises AD patients and healthy subjects; wherein the validated second selected SGV combinations are synergistic in separating an AD patient or subject at risk of developing AD from a healthy subject. In one embodiment thereto, the selection was performed by an unbiased analysis searching for combinations of the 83 Alzheimer disease (AD) genome-wide association study (GWAS) significant short genetic variants (SGVs, pathogenic SGVs) that separate AD patients from controls (listed in Table 1 herein; Bellenguez et al. 2022, Nature Genetics 54:412-436). The majority of SGVs are single nucleotide polymorphisms (SGVs) except for 4 SGVs characterized by short deletions or deletions, i.e. the SGVs in lines ("#") 32, 33, 41, and 59 of Table 1. In view of these single SGVs identified by Bellenguez et al. 2022 being able to separate / distinguish AD patients or subjects at risk of developing AD from healthy / control subjects, these SGVs, and more in particular the SGV variants containing the nucleotide(s) of the risk or pathogenic allele (i.e. the effect allele with a beta coefficient value >0, or the other allele in case the effect allele has a beta coefficient value <0, see Table 1) are also termed "risk SGVs" or "pathogenic SGVs", and are interchangeably referred to herein sometimes simply as "variants" or "polymorphisms". The corresponding risk or pathogenic nucleotides / risk or pathogenic alleles of a pathogenic SGV are separately listed in Table 1. This set of 83 AD-associated SGVs is one exemplary set of single SGVs that can be used in the above-described method of / for selecting SGV combinations.

[0027] In a further embodiment to any of the above, the MDR analysis can be run on more than 1 training set of subjects, e.g. on 2, 3, 4, 5 or more training sets of subjects, more in particular on 2 training sets of subjects. More in particular, these training sets are selected such that the distribution of subjects and / or subject characteristics (sex, age, etc.) are similar for each of the individual training sets. More in particular, such more than 1 training sets can be composed in a computer assisted manner, e.g. using a caret algorithm (Classification And REgression Training algorithm).

[0028] Furthermore in particular, when using more than 1 training set of subjects, the MDR analysis will provide a set of first selected SGV combinations for each of the individual training sets (i.e. more than 1 set of first selected SGV combinations, of which 1 set per training set of subjects). As additional step in the above-described methods, only those first selected SGV combinations common to or shared by all used training sets are selected for the next analysis step. Thus, when 2 training sets are used, the first selected SGV combinations are those SGV combinations occurring in both of / common to / shared by the 2 training sets.

[0029] Furthermore in particular, other covariates or confounding factors (other than the main effects of the single SGVs) are incorporated in the analysis, more particular in the logistic regression analysis step. Such other covariates include one or more of sex, age, APOE status, and genetic principal components (gPCs) distilled from the cohort of subjects as used for the analysis. Such covariates / confounding factors can be selected by e.g. running a likelihood-ratio test, e.g. with the base R function Irtest (likelihood ratio test of nested models).

[0030] Furthermore in particular, the validation is based on the ability of an SGV combination to increase Area Under the Curve (AUC) when computing the Receiver Operator Characteristic curve (ROC) relative to an additive model constructed with the 85 different individual SGV variants (including APOE SGVs rs429358 and APOE SGV rs7412).

[0031] Furthermore in particular, only those combinations that are predicted to have a pathogenic epistatic effect (beta coefficient > 0) and are significant (p <0.05) are selected.

[0032] Furthermore in particular, only those combinations that, individually, increase the prediction ability of the additive model constructed with the 85 SGVs , selected for the final stage of the construction of the prediction epistatic genetic risk score model.

[0033] Thus, the above computer implemented methods or computer programs to perform methods of selecting single genetic variant (SGV) combinations can comprise alternative steps such as: a first selection of SGV combinations by multifactor dimensionality reduction (MDR) analysis on two independent training sets of subjects wherein the training set comprises AD patients and healthy / control subjects; selection of the SGV combinations that are replicated between the two training sets or are present or represented in the two training sets; and / or such as: logistic regression (LR) analysis of the first selected SGV combinations wherein the logistic regression includes the main effect of the single SGVs, age, sex and a set of genetic principal components (e.g. the 20 first or top 20 gPCs) as covariates; and / or such as: validating the second selected SGV combinations in a test set of subjects wherein the test set comprises AD patients and healthy subjects based on their ability to increase Area Under the Curve (AUC) when computing the Receiver Operator Characteristic curve (ROC) when combined to an additive model constructed with a set of different pathogenic SGVs (e.g. the SGV set of Table 1, further including APOE SGVs rs429358 and APOE SGV rs7412).

[0034] In a further aspect thereto, this disclosure relates to a computer product comprising a computer readable medium storing instructions for operating a computer system to perform at least one analysis and / or calculation as outlined hereinabove. In particular, the at least one analysis and / or calculation is a multifactor dimensionality reduction analysis and / or a logistic regression analysis. Alternatively envisaged are computing or data processing systems or devices, machine readable media, or diagnostic kits comprising a means for carrying out or performing a method, or for carrying out or performing a step of a method as described above.

[0035] In a further aspect of this disclosure, SGV combinations of interest were identified. Combinations of 2 SGVs are listed in Tables 3A / B and Tables 4A / B. Combinations of 3 SGVs are listed in Tables 5A / B and 6A / B. Combinations of 4 SGVs are listed in Tables 7A / B and 8A / B. Alternative combinations of 2 SGVs (based on prior segregation / stratification of individuals in a dataset according to number of APOEE4 alleles / according to APOEE4 background) are listed in Tables 9A, IDA and 11A. All of these individual SGV combinations are part of this disclosure, as well as sets (more than one) of these individual SGV combinations. In view of these (sets of) SGV combinations being able to diagnose / determine / assess / prognosticate / identify a subject as AD patient or as potential AD patient, or as subject at risk of developing AD, or to separate AD patients or subjects at risk of developing AD from healthy subjects, these (sets of) SGVs are also termed "(sets of) risk (SGV) combinations", "(sets of) pathogenic (SGV) combinations" or "(sets of) AD risk (SGV) combinations".

[0036] A further aspect of this disclosure relates to methods of diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or potential AD patient, or as subject at risk of developing AD, such methods including one or more steps of: determining the presence or absence of a pathogenic allele of the individual short genetic variants (SGVs) of a SGV combination in a DNA sample obtained from a subject, wherein the SGV combination is chosen from the SGV combinations as listed / given in Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table IDA or Table 11A; diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or as potential AD patient, or as subject at risk of developing AD, when a pathogenic allele is present in each of the individual SGVs of the SGV combination.

[0037] Alternatively, such methods are one or more steps of: determining the presence or absence of a pathogenic allele of the individual short genetic variants (SGVs) of more than one / at least two SGV combinations in a DNA sample obtained from a subject, wherein the SGV combinations are chosen from the SGV combinations as listed / given in Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table IDA or Table 11A; diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or as potential AD patient, or as subject at risk of developing AD, when a pathogenic allele is present in each of the individual SGVs of the more than one / at least two SGV combinations.

[0038] Alternatively, such methods are one or more steps of: determining the nucleotides or nature of the nucleotides in the alleles of the individual SGVs of a SGV combination in a DNA sample obtained from a subject, or determining the nucleotide sequences of the alleles (or of the loci comprising the alleles) of the individual SGVs of a SGV combination chosen from or as listed / given in Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table IDA orTable 11A, in a DNA sample obtained from a subject; d iagn osi ng / d eterm i n i ng / assessing / progn osticati ng / identifying a subject as AD patient or as potential AD patient, or as subject at risk of developing AD, when the nucleotide in an allele of each of the individual SGVs of the SGV combination is a pathogenic nucleotide, or when a pathogenic allele is present in each of the individual SGVs of the SGV combination.

[0039] Alternatively, such methods are one or more steps of: determining the nucleotides or nature of the nucleotides in the alleles of the individual SGVs of more than one / at least two SGV combinations in a DNA sample obtained from a subject, or determining the nucleotide sequences of the alleles (or of the loci comprising the alleles) of the individual SGVs of more than one / at least two SGV combinations, wherein the SGV combinations are chosen from the SGV combinations listed / given in Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table IDA or Table 11A, in a DNA sample obtained from a subject; diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or as potential AD patient, or as subject at risk of developing AD, when the nucleotide in an allele of each of the individual SGVs of the more than one / at least two SGV combinations is a pathogenic nucleotide, or when a pathogenic allele is present in each of the individual SGVs of the more than one / at least two SGV combinations.

[0040] Alternatively, such methods are including one or more steps of: determining the number of pathogenic alleles of the individual short genetic variants (SGVs) of a SGV combination in a DNA sample obtained from a subject, wherein the SGV combination is chosen from the SGV combinations as listed / given in Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table IDA or Table 11A; diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or as potential AD patient, or as subject at risk of developing AD, when in each of the individual SGVs of the SGV combination the number of pathogenic alleles as listed for the SGV combination chosen from Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table IDA or Table 11A, is present.

[0041] Alternatively, such methods are including one or more steps of: determining the number of pathogenic alleles of the individual short genetic variants (SGVs) of more than one / at least two SGV combinations in a DNA sample obtained from a subject, wherein the more than one / at least two SGV combinations are chosen from the SGV combinations as listed / given in Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table IDA or Table 11A; diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or as potential AD patient, or as subject at risk of developing AD, when in each of the individual SGVs of the more than one / at least two SGV combinations the number of pathogenic alleles as listed for the SGV combinations chosen from Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table IDA or Table 11A, is present.

[0042] Alternatively, the above methods of diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or as subject at risk of developing AD are:

[0043] - methods of diagnosing / determining / assessing / prognosticating / identifying a subject to have AD or to potentially have AD, or to be at risk of developing AD; in particular the step of "diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or as potential AD patient, or as subject at risk of developing AD" is then substituted for a step of "diagnosing / determining / assessing / prognosticating / identifying a subject to have AD or to potentially have AD, or to be at risk of developing AD"; or

[0044] - methods of separating or distinguishing AD patients or subjects at risk of developing AD from healthy or control subject; in particular the step of

[0045] "diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or as potential AD patient, or as subject at risk of developing AD" is then substituted for a step of "separating or distinguishing AD patients or subjects at risk of developing AD from healthy or control subject" based on the presence of the specified pathogenic allele(s) or pathogenic nucleotide(s).

[0046] In relation to these methods, the pathogenic alleles / pathogenic nucleotides of the individual SGVs are more particular listed / given in Table 1.

[0047] Furthermore in particular in relation to these methods, these may relate to multiplex analysis, more particular multiplex determination of the presence or absence of pathogenic alleles, multiplex determination of the number of pathogenic alleles, multiplex determination of the (nature of the) nucleotide in the alleles, or multiplex determination of the nucleotide sequences of the alleles (or of the loci comprising the alleles).

[0048] With "locus comprising an allele" is meant in the present context the genetic locus corresponding to those comprising or spanning a single SGV (as included in a SGV combination), or the locus of an individual SGVs of the SGV combination. Taking, by way of example, the SGV combination of line #1 of Table 3A, i.e. the combination of SGV TREM2 rs75932628 and SGV ABCA7 rsl2151021. From Table 1, it can be derived that SGV TREM2 rs75932628 is located on chromosome 6 at position 41161514 ("chr6 41161514"), and this position can be a T or C nucleotide (the nature of the nucleotide in the variant position can thus be T or C); for the SGV ABCA7 rsl2151021, the locus is located at position 1050875 on chromosome 19 ("chrl9 1050875") and this position can be an A or G nucleotide (the nature of the nucleotide in the variant position can thus be A or G) [locations according to the GRCh / hg38 reference genome]. If from allele or sequence analysis of the DNA from a subject it is obvious that the DNA comprises a T nucleotide at position chr641161514 and an A nucleotide at position chrl9 1050875, then the subject is comprising a pathogenic nucleotide / allele of SGV TREM2 rs75932628 and a pathogenic nucleotide / allele of SGV ABCA7 rsl2151021, and is thus comprising a pathogenic SGV combination.

[0049] In one embodiment thereto, the presence of an SGV of interest (pathogenic SGV), on a single allele is sufficient. In a further embodiment, an SGV of interest is present on both alleles. As such, different combinations of SGVs with allele number can arise. In case of the specific example of the combination of SGVTREM2 rs75932628 and SGV ABCA7 rsl2151021, the pathogenic allele of SGVTREM2 rs75932628 is present on one (1) allele, and the pathogenic allele of SGV ABCA7 rsl2151021 is present on two (2) / both alleles (Table 3A / B). This same SGV combination can in principle be similarly pathogenic in other allele combinations (1 / 1, 2 / 1, 1 / 2, 2 / 2). Such other allele combinations may, however, not have been present in the cohort of subjects at the basis of the selection procedure.

[0050] With "more than one / at least two SGV combinations" or "more than one / at least two pathogenic SGV combinations" is meant a set of more than one / at least two SGV combinations or pathogenic SGV combinations. More in particular, each individual SGV combination of such set is selected from the SGV combinations as listed / given in Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table IDA and / or Table 11A. More in particular, such sets are sets of 2, 3, 4, 5, 6, 7, 8, 9, 10 or more SGV combinations chosen from Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table 10A and / or Table 11A; or sets of 2 to 20, 2 to 25, 2 to 30, 2 to 35, 2 to 40, 2 to 45, 2 to 50, or 2 to 100 SGV combinations chosen from Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table 10A and / or Table 11A.

[0051] In further embodiments to these methods of / for separating / distinguishing AD patients or subjects at risk of developing AD from healthy / control subjects, and to these methods of / for diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or as potential AD patient, or as subject at risk of developing AD, the single (pathogenic) SGV combination may in particular be one of those listed / given in the paragraphs hereinafter (paragraphs starting with "As such, in particular"); or the set of more than one / at least two (pathogenic) SGV combinations may in particular comprise one SGV combination of those listed / given in the paragraphs hereinafter (paragraphs starting with "As such, in particular").

[0052] Selected combinations of 2 SGVs identified without prior stratification of the training dataset according to number of APOEE4 alleles / according to APOEE4 background are listed hereinafter. As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs including or excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 5, i.e. the combinations listed in Table 3B (in lines #1 to and including #18) and in Table 4B (in lines #1 to and including #19). More in particular, these are the combinations of the SGV pairs:

[0053] MME rs61762319 and TSPOAP1 rs2526377 (occurring in lines #1 and 3 of Table 3B and Table 4B in different combinations of number of risk alleles per SGV);

[0054] BLNK rs6584063 and SLC24A4 rsl2590654 (occurring in lines #2 and 5 of Table 3B and Table 4B in different combinations of number of risk alleles per SGV);

[0055] SORL1 rsll218343 and TPCN1 rs6489896 (occurring in lines #4 and 6 of Table 3B and Table 4B in different combinations of number of risk alleles per SGV);

[0056] MME rsl6824536 and TPCN1 rs6489896 (occurring in lines #7 and 11 of Table 3B and Table 4B in different combinations of number of risk alleles per SGV);

[0057] MME rsl6824536 and HS3ST5 rs785129 (occurring in lines #8 and 10 of Table 3B and Table 4B in different combinations of number of risk alleles per SGV);

[0058] CTSB rsl065712 and LILRB2 rs587709 (occurring in line #9 of Table 3B and Table 4B);

[0059] BLNK rs6584063 and MS4A4A rsl582763 (occurring in lines #12, 14 and 15 of Table 3B and Table 4B in different combinations of number of risk alleles per SGV);

[0060] TREM2 rs75932628 and ABCA7 rsl2151021 (occurring in line #13 of Table 3B and Table 4B);

[0061] SORL1 rsll218343 and ACE rs4277405 (occurring in line #16 of Table 3B and in line #17 of Table 4B);

[0062] TREML2 rs60755019 and BCKDK rs889555 (occurring in lines #18 and 19 of Table 3B and in line #18 of Table 4B).

[0063] Such further combination including the SGV APOE rs429358 is the combination of MINDY2 rs602602 and APOE rs7412 (occurring in line #17 of Table 3B).

[0064] Such further combinations excluding the SGV APOE rs429358 are the combinations of ANKH rsll2403360 and CTSB rsl065712 (occurring in line #16 of Table 4B) and SORL1 rsll218343 and APP rs2154481 (occurring in line #19 of Table 4B).

[0065] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs including or excluding the SGV APOE rs429358 are combinations with an odds ratio equal to or higher than 10, i.e. the combinations listed in Table 3B and in Table 4B, in lines #1 to and including #12 in both. More in particular, these are the combinations of the SGV pairs:

[0066] MME rs61762319 and TSPOAP1 rs2526377 (occurring in lines #1 and 3 of Table 3B and Table 4B in different combinations of number of risk alleles per SGV);

[0067] BLNK rs6584063 and SLC24A4 rsl2590654 (occurring in lines #2 and 5 of Table 3B and Table 4B in different combinations of number of risk alleles per SGV); S0RL1 rsll218343 and TPCN1 rs6489896 (occurring in lines #4 and 6 of Table 3B and Table 4B in different combinations of number of risk alleles per SGV);

[0068] MME rsl6824536 and TPCN1 rs6489896 (occurring in lines #7 and 11 of Table 3B and Table 4B in different combinations of number of risk alleles per SGV);

[0069] MME rsl6824536 and HS3ST5 rs785129 (occurring in lines #8 and 10 of Table 3B and Table 4B in different combinations of number of risk alleles per SGV);

[0070] CTSB rsl065712 and LILRB2 rs587709 (occurring in line #9 of Table 3B and Table 4B);

[0071] BLNK rs6584063 and MS4A4A rsl582763 (occurring in line #12 of Table 3B and Table 4B).

[0072] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs including or excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 15, i.e. the combinations listed in Table 3B and Table 4B, in lines #1 to and including #7 in both. More in particular, these are the combinations of the SGV pairs:

[0073] MME rs61762319 and TSPOAP1 rs2526377 (occurring in lines #1 and 3 of Table 4B and Table 5B in different combinations of number of risk alleles per SGV);

[0074] BLNK rs6584063 and SLC24A4 rsl2590654 (occurring in lines #2 and 5 of Table 4B and Table 5B in different combinations of number of risk alleles per SGV);

[0075] SORL1 rsll218343 and TPCN1 rs6489896 (occurring in lines #4 and 6 of Table 4B and Table 5B in different combinations of number of risk alleles per SGV);

[0076] MME rsl6824536 and TPCN1 rs6489896 (occurring in line #7 of Table 3B and Table 4B).

[0077] As such, in particular, a combination of 2 SGVs is a combination of the SGV MME rs61762319 and SGV TSPOAP1 rs2526377 (occurring in lines #1 and 3 of Table 4B and Table 5B). More in particular in this combination, both alleles of the subject of interest have the SGV MME rs61762319 and the SGVTSPOAP1 rs2526377 (homo-allelic SGVs).

[0078] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs including the SGV APOE rs429358 is a combination comprising one SGV with a frequency of occurrence (Table 3C) of at least 9 out of the 268 withheld combinations. More in particular a combination of 2 SGVs is a combination of 2 SGVs comprising at least one of the SGVs chosen from: WNT3 rsl99515, UNC5CL rsl0947943, CLNK rs6846529, ABCA7 rsl2151021, APP rs2154481, COX7C rs62374257, CTSH rsl2592898, MY015A rs2242595, TSPAN14 rs6586028, UMAD1 rs6943429, IGHcluster rsl0131280, APOE rs429358, CD2AP rs7767350, CR1 rs679515, EED rs3851179, ICA1 rsl0952097, INPP5D rsl0933431, MS4A4A rsl582763, SPI1 rsl0437655, PLCG2 rsl2446759, RASGEF1C rsll3706587, RHOH rs2245466, and TSP0AP1 rs2526377. In one embodiment thereto, the chosen SGV is combined with a second SGV of Table 1 different from the chosen SGV.

[0079] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs including the SGV APOE rs429358 is a combination comprising one SGV with a frequency of occurrence (Table 3C) of at least 10 out of the 268 withheld combinations. More in particular a combination of 2 SGVs is a combination of 2 SGVs comprising at least one of the SGVs chosen from: WNT3 rsl99515, UNC5CL rsl0947943, CLNK rs6846529, ABCA7 rsl2151021, APP rs2154481, COX7C rs62374257, CTSH rsl2592898, MYO15A rs2242595, TSPAN14 rs6586028, UMAD1 rs6943429, IGHcluster rsl0131280, APOE rs429358, CD2AP rs7767350, CR1 rs679515, EED rs3851179, ICA1 rsl0952097, INPP5D rsl0933431, MS4A4A rsl582763, and SPI1 rsl0437655. In one embodiment thereto, the chosen SGV is combined with a second SGV of Table 1 different from the chosen SGV.

[0080] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs including the SGV APOE rs429358 is a combination comprising one SGV with a frequency of occurrence (Table 3C) of at least 11 out of the 268 withheld combinations. More in particular a combination of 2 SGVs is a combination of 2 SGVs comprising at least one of the SGVs chosen from: WNT3 rsl99515, UNC5CL rsl0947943, CLNK rs6846529, ABCA7 rsl2151021, APP rs2154481, COX7C rs62374257, CTSH rsl2592898, MYO15A rs2242595, TSPAN14 rs6586028, UMAD1 rs6943429, and IGHcluster rsl0131280. In one embodiment thereto, the chosen SGV is combined with a second SGV of Table 1 different from the chosen SGV.

[0081] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs including the SGV APOE rs429358 is a combination comprising one SGV with a frequency of occurrence (Table 3C) of at least 12 out of the 268 withheld combinations. More in particular a combination of 2 SGVs is a combination of 2 SGVs comprising at least one of the SGVs chosen from: WNT3 rsl99515, UNC5CL rsl0947943, CLNK rs6846529, ABCA7 rsl2151021, APP rs2154481, COX7C rs62374257, CTSH rsl2592898, MYO15A rs2242595, TSPAN14 rs6586028, and UMAD1 rs6943429. In one embodiment thereto, the chosen SGV is combined with a second SGV of Table 1 different from the chosen SGV.

[0082] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs including the SGV APOE rs429358 is a combination comprising one SGV with a frequency of occurrence (Table 3C) of at least 13 out of the 268 withheld combinations. More in particular a combination of 2 SGVs is a combination of 2 SGVs comprising at least one of the SGVs chosen from: WNT3 rsl99515, UNC5CL rsl0947943, CLNK rs6846529, ABCA7 rsl2151021, and APP rs2154481. In one embodiment thereto, the chosen SGV is combined with a second SGV of Table 1 different from the chosen SGV.

[0083] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs including the SGV APOE rs429358 is a combination comprising one SGV with a frequency of occurrence (Table 3C) of at least 14 out of the 268 withheld combinations. More in particular a combination of 2 SGVs is a combination of 2 SGVs comprising at least one of the SGVs chosen from: WNT3 rsl99515, UNC5CL rsl0947943, and CLNK rs6846529. In one embodiment thereto, the chosen SGV is combined with a second SGV of Table 1 different from the chosen SGV.

[0084] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs including the SGV APOE rs429358 is a combination comprising one SGV with a frequency of occurrence (Table 3C) of at least 15 out of the 268 withheld combinations. More in particular a combination of 2 SGVs is a combination of 2 SGVs comprising at least one of the SGVs chosen from: WNT3 rsl99515, and UNC5CL rsl0947943. In one embodiment thereto, the chosen SGV is combined with a second SGV of Table 1 different from the chosen SGV.

[0085] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs including the SGV APOE rs429358 is a combination comprising the SGV UNC5CL rsl0947943. More in particular a combination of 2 SGVs is a combination of 2 SGVs comprising the SGV UNC5CL rsl0947943. In one embodiment thereto, the SGV UNC5CL rsl0947943 is combined with a second SGV of Table 1 different from the SGV UNC5CL rsl0947943.

[0086] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs including the SGV APOE rs429358 is a combination comprising the SGV WNT3 rsl99515. More in particular a combination of 2 SGVs is a combination of 2 SGVs comprising the SGV WNT3 rsl99515. In one embodiment thereto, the SGV WNT3 rsl99515 is combined with a second SGV of Table 1 different from the SGV WNT3 rsl99515.

[0087] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs excluding the SGV APOE rs429358 is a combination comprising one SGV with a frequency of occurrence (Table 4C) of at least 8 out of the 264 withheld combinations. More in particular a combination of 2 SGVs is a combination of 2 SGVs comprising at least one of the SGVs chosen from: WNT3 rsl99515, UNC5CL rsl0947943, CLNK rs6846529, ABCA7 rsl2151021, APP rs2154481, COX7C rs62374257, CTSH rsl2592898, MY015A rs2242595, TSPAN14 rs6586028, UMAD1 rs6943429, ICA1 rsl0952097, IGHcluster rsl0131280, CD2AP rs7767350, CR1 rs679515, EED rs3851179, INPP5D rsl0933431, PLCG2 rsl2446759, RHOH rs2245466, SPI1 rsl0437655, RASGEF1C rsll3706587, TSPOAP1 rs2526377, CASS4 rs6014724, DOC2A rsll40239, HS3ST5 rs785129, MS4A4A rsl582763, RBCK1 rsl358782, and SNX1 rs3848143. In one embodiment thereto, the chosen SGV is combined with a second SGV of Table 1 different from the chosen SGV.

[0088] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs excluding the SGV APOE rs429358 is a combination comprising one SGV with a frequency of occurrence (Table 4C) of at least 9 out of the 264 withheld combinations. More in particular a combination of 2 SGVs is a combination of 2 SGVs comprising at least one of the SGVs chosen from: WNT3 rsl99515, UNC5CL rsl0947943, CLNK rs6846529, ABCA7 rsl2151021, APP rs2154481, COX7C rs62374257, CTSH rsl2592898, MY015A rs2242595, TSPAN14 rs6586028, UMAD1 rs6943429, ICA1 rsl0952097, IGHcluster rsl0131280, CD2AP rs7767350, CR1 rs679515, EED rs3851179, INPP5D rsl0933431, PLCG2 rsl2446759, RHOH rs2245466, SPI1 rsl0437655, RASGEF1C rsll3706587, and TSPOAP1 rs2526377. In one embodiment thereto, the chosen SGV is combined with a second SGV of Table 1 different from the chosen SGV.

[0089] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs excluding the SGV APOE rs429358 is a combination comprising one SGV with a frequency of occurrence (Table 4C) of at least 10 out of the 264 withheld combinations. More in particular a combination of 2 SGVs is a combination of 2 SGVs comprising at least one of the SGVs chosen from: WNT3 rsl99515, UNC5CL rsl0947943, CLNK rs6846529, ABCA7 rsl2151021, APP rs2154481, COX7C rs62374257, CTSH rsl2592898, MYO15A rs2242595, TSPAN14 rs6586028, UMAD1 rs6943429, ICA1 rsl0952097, IGHcluster rsl0131280, CD2AP rs7767350, CR1 rs679515, EED rs3851179, INPP5D rsl0933431, PLCG2 rsl2446759, RHOH rs2245466, and SPI1 rsl0437655. In one embodiment thereto, the chosen SGV is combined with a second SGV of Table 1 different from the chosen SGV.

[0090] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs excluding the SGV APOE rs429358 is a combination comprising one SGV with a frequency of occurrence (Table 4C) of at least 11 out of the 264 withheld combinations. More in particular a combination of 2 SGVs is a combination of 2 SGVs comprising at least one of the SGVs chosen from: WNT3 rsl99515, UNC5CL rsl0947943, CLNK rs6846529, ABCA7 rsl2151021, APP rs2154481, COX7C rs62374257, CTSH rsl2592898, MYO15A rs2242595, TSPAN14 rs6586028, UMAD1 rs6943429, ICA1 rsl0952097, and IGHcluster rsl0131280. In one embodiment thereto, the chosen SGV is combined with a second SGV of Table 1 different from the chosen SGV. As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs excluding the SGV APOE rs429358 is a combination comprising one SGV with a frequency of occurrence (Table 4C) of at least 12 out of the 264 withheld combinations. More in particular a combination of 2 SGVs is a combination of 2 SGVs comprising at least one of the SGVs chosen from: WNT3 rsl99515, UNC5CL rsl0947943, CLNK rs6846529, ABCA7 rsl2151021, APP rs2154481, COX7C rs62374257, CTSH rsl2592898, MY015A rs2242595, TSPAN14 rs6586028, and UMAD1 rs6943429. In one embodiment thereto, the chosen SGV is combined with a second SGV of Table 1 different from the chosen SGV.

[0091] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs excluding the SGV APOE rs429358 is a combination comprising one SGV with a frequency of occurrence (Table 4C) of at least 13 out of the 264 withheld combinations. More in particular a combination of 2 SGVs is a combination of 2 SGVs comprising at least one of the SGVs chosen from: WNT3 rsl99515, UNC5CL rsl0947943, CLNK rs6846529, ABCA7 rsl2151021, and APP rs2154481. In one embodiment thereto, the chosen SGV is combined with a second SGV of Table 1 different from the chosen SGV.

[0092] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs excluding the SGV APOE rs429358 is a combination comprising one SGV with a frequency of occurrence (Table 4C) of at least 14 out of the 264 withheld combinations. More in particular a combination of 2 SGVs is a combination of 2 SGVs comprising at least one of the SGVs chosen from: WNT3 rsl99515, UNC5CL rsl0947943, and CLNK rs6846529. In one embodiment thereto, the chosen SGV is combined with a second SGV of Table 1 different from the chosen SGV.

[0093] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs excluding the SGV APOE rs429358 is a combination comprising the SGV UNC5CL rsl0947943. More in particular a combination of 2 SGVs is a combination of 2 SGVs comprising the UNC5CL rsl0947943. In one embodiment thereto, the SGV UNC5CL rsl0947943 is combined with a second SGV of Table 1 different from the SGV UNC5CL rsl0947943.

[0094] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs excluding the SGV APOE rs429358 is a combination comprising the SGV WNT3 rsl99515. More in particular a combination of 2 SGVs is a combination of 2 SGVs comprising the SGV WNT3 rsl99515. In one embodiment thereto, the SGV WNT3 rsl99515 is combined with a second SGV of Table 1 different from the SGV WNT3 rsl99515. In an alternative manner of determining combinations of SGVs that are significantly synergistic in distinguishing AD versus non-AD, the training dataset (EADB training dataset) was subdivided into 3 main subgroups: a subgroup corresponding to individuals carrying or containing zero alleles of APOEE4 (APOEE4 - / - group) / a subgroup corresponding to individuals carrying or containing 1 allele of APOEE4 (APOEE4 + / - group) and a subgroup corresponding to individuals carrying or containing APOEE4 (APOEE4 + / + group). In each of those dataset subgroup, the same analysis was then performed as when identifying combinations of SGVs without stratifying the training dataset according to number of APOEE4 alleles / according to APOEE4 background. The combinations of 2 SGVs identified this way are listed in Table 9A (APOEE4 - / - group), Table IDA (APOEE4 + / - group), and Table 11A (APOEE4 + / + group).

[0095] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs based on the subset of subjects not carrying the SGV APOE rs429358 allele (which encodes for the pathogenic APOEE4 isoform; APOEE4 - / - group), is a combination with an odds ratio equal to or higher than 5, i.e. the combinations listed in Table 9A (in lines #1 to and including #27). More in particular, these are the combinations of the SGV pairs:

[0096] MME rsl6824536 and MS4A4A rsl582763 (occurring in lines #1 and 2 of Table 9A in different combinations of number of risk alleles per SGV);

[0097] SHARPIN rs34173062 and FOXF1 rsl6941239 (occurring in line #3 of Table 9A);

[0098] COX7C rs62374257 and BLNK rs6584063 (occurring in lines #4 and 6 of Table 9A in different combinations of number of risk alleles per SGV);

[0099] TPCN1 rs6489896 and BCKDK rs889555 (occurring in lines #5 and 12 of Table 9A in different combinations of number of risk alleles per SGV);

[0100] TREML2 rs60755019 and BCKDK rs889555 (occurring in lines #7 and 8 of Table 9A in different combinations of number of risk alleles per SGV);

[0101] BLNK rs6584063 and SLC24A4 rsl2590654 (occurring in line #9 of Table 9A);

[0102] BLNK rs6584063 and KLF16 rsl49080927 (occurring in lines #10 and 13 of Table 9A in different combinations of number of risk alleles per SGV);

[0103] UNC5CL rsl0947943 and SORL1 rsll218343 (occurring in lines #11 and 23 of Table 9A in different combinations of number of risk alleles per SGV);

[0104] TSPAN14 rs6586028 and CTSH rsl2592898 (occurring in lines #14, 15, 17 and 18 of Table 9A in different combinations of number of risk alleles per SGV);

[0105] RHOH rs2245466 and SORL1 rsll218343 (occurring in lines #16 and 19 of Table 9A in different combinations of number of risk alleles per SGV);

[0106] ANKH rsll2403360 and SPPL2A rs8025980 (occurring in line #20 of Table 9A);

[0107] TREM2 rs75932628 and ABCA7 rsl2151021 (occurring in line #21 of Table 9A); S0RT1 rsl41749679 and CLNK rs6846529 (occurring in line #22 of Table 9A);

[0108] ICA1 rsl0952097 and SPDYE3 rs7384878 (occurring in line #23 of Table 9A);

[0109] ICA1 rsl0952097 and SPDYE3 rs7384878 (occurring in line #24 of Table 9A);

[0110] SEC61G rs76928645 and SLC2A4RG rs6742 (occurring in line #25 of Table 9A);

[0111] HLAdqal rs6605556 and APH1B rsll7618017 (occurring in line #26 of Table 9A);

[0112] HLAdqal rs6605556 and TSPAN14 rs6586028 (occurring in line #27 of Table 9A).

[0113] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs based on the subset of subjects not carrying the SGV APOE rs429358 allele (which encodes for the pathogenic APOEE4 isoform; APOEE4 - / - group), is a combination with an odds ratio equal to or higher than 10, i.e. the combinations listed in Table 9A (in lines #1 to and including #12). More in particular, these are the combinations of the SGV pairs:

[0114] MME rsl6824536 and MS4A4A rsl582763 (occurring in lines #1 and 2 of Table 9A in different combinations of number of risk alleles per SGV);

[0115] SHARPIN rs34173062 and FOXF1 rsl6941239 (occurring in line #3 of Table 9A);

[0116] COX7C rs62374257 and BLNK rs6584063 (occurring in lines #4 and 6 of Table 9A in different combinations of number of risk alleles per SGV);

[0117] TPCN1 rs6489896 and BCKDK rs889555 (occurring in lines #5 and 12 of Table 9A in different combinations of number of risk alleles per SGV);

[0118] TREML2 rs60755019 and BCKDK rs889555 (occurring in lines #7 and 8 of Table 9A in different combinations of number of risk alleles per SGV);

[0119] BLNK rs6584063 and SLC24A4 rsl2590654 (occurring in line #9 of Table 9A);

[0120] BLNK rs6584063 and KLF16 rsl49080927 (occurring in line #10 of Table 9A);

[0121] UNC5CL rsl0947943 and SORL1 rsll218343 (occurring in line #11 of Table 9A).

[0122] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs based on the subset of subjects not carrying the SGV APOE rs429358 allele (which encodes for the pathogenic APOEE4 isoform; APOEE4 - / - group), is a combination with an odds ratio equal to or higher than 15, i.e. the combinations listed in Table 9A (in lines #1 to and including #2). More in particular, these are the combinations of the SGV pairs:

[0123] MME rsl6824536 and MS4A4A rsl582763 (occurring in lines #1 and 2 of Table 9A in different combinations of number of risk alleles per SGV).

[0124] MME rsl6824536 and MS4A4A rsl582763 Furthermore in particular, the combination of 2 SGVs is MME rsl6824536 and MS4A4A rsl582763. Furthermore in particular hereto MS4A4A rsl582763 is occurring in 1 allele whereas MME rsl6824536 is occurring in 1 or 2 alleles.

[0125] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs based on the subset of subjects carrying a single SGV APOE rs429358 allele (which encodes for the pathogenic APOEE4 isoform; APOEE4 + / - group), is a combination with an odds ratio equal to or higher than 5, i.e. the combinations listed in Table IDA (in lines #1 to and including #21). More in particular, these are the combinations of the SGV pairs:

[0126] WDR12 rsl39643391 and ADAMTS1 rs2830489 (occurring in lines #1 to 4 of Table IDA in different combinations of number of risk alleles per SGV);

[0127] JAZF1 rsll60871 and SCIMP rs7225151 (occurring in line #5 of Table IDA);

[0128] SORL1 rsll218343 and CTSH rsl2592898 (occurring in line #6 of Table IDA);

[0129] TREM2 rsl43332484 and SEC61G rs76928645 (occurring in line #7 of Table IDA);

[0130] MME rsl6824536 and HS3ST5 rs785129 (occurring in lines #8 and 10 of Table 10A in different combinations of number of risk alleles per SGV);

[0131] JAZF1 rsll60871 and SCIMP rs7225151 (occurring in line #9 of Table 10A);

[0132] MME rsl6824536 and UMAD1 rs6943429 (occurring in lines #11 and 12 of Table 10A in different combinations of number of risk alleles per SGV);

[0133] SHARPIN rs34173062 and SLC24A4 rs7401792 (occurring in line #13 of Table 10A);

[0134] USP6NL rs7912495 and CASS4 rs6014724 (occurring in lines #14 and 18 of Table 10A in different combinations of number of risk alleles per SGV);

[0135] NCK2 rsl43080277 and SLC24A4 rsl2590654 (occurring in lines #15 and 20 of Table 10A in different combinations of number of risk alleles per SGV);

[0136] CR1 rs679515 and MY015A rs2242595 (occurring in line #16 of Table 10A);

[0137] SCIMP rs7225151 and RBCK1 rsl358782 (occurring in line #17 of Table 10A);

[0138] SLC24A4 rsl2590654 and APH1B rsll7618017 (occurring in line #19 of Table 10A);

[0139] ANKH rsll2403360 and PLEKHA1 rs7908662 (occurring in line #21 of Table 10A).

[0140] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs based on the subset of subjects carrying a single SGV APOE rs429358 allele (which encodes for the pathogenic APOEE4 isoform; APOEE4 + / - group), is a combination with an odds ratio equal to or higher than 10, i.e. the combinations listed in Table 10A (in lines #1 to and including #10). More in particular, these are the combinations of the SGV pairs: WDR12 rsl39643391 and ADAMTS1 rs2830489 (occurring in lines #1 to 4 of Table IDA in different combinations of number of risk alleles per SGV);

[0141] JAZF1 rsll60871 and SCIMP rs7225151 (occurring in line #5 of Table IDA);

[0142] SORL1 rsll218343 and CTSH rsl2592898 (occurring in line #6 of Table IDA);

[0143] TREM2 rsl43332484 and SEC61G rs76928645 (occurring in line #7 of Table IDA);

[0144] MME rsl6824536 and HS3ST5 rs785129 (occurring in line #8 of Table IDA);

[0145] JAZF1 rsll60871 and SCIMP rs7225151 (occurring in line #9 of Table IDA); MME rsl6824536 and HS3ST5 rs785129 (occurring in line #10 of Table 10A).

[0146] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs based on the subset of subjects carrying a single SGV APOE rs429358 allele (which encodes for the pathogenic APOEE4 isoform; APOEE4 + / - group), is a combination with an odds ratio equal to or higher than 15, i.e. the combinations listed in Table 10A (in lines #1 to and including #5). More in particular, these are the combinations of the SGV pairs:

[0147] WDR12 rsl39643391 and ADAMTS1 rs2830489 (occurring in lines #1 to 4 of Table IDA in different combinations of number of risk alleles per SGV);

[0148] JAZF1 rsll60871 and SCIMP rs7225151 (occurring in line #5 of Table IDA).

[0149] Furthermore in particular, the combination of 2 SGVs is WDR12 rsl39643391 and ADAMTS1 rs2830489. Furthermore in particular hereto WDR12 rsl39643391 and ADAMTS1 rs2830489 are both occurring in 2 alleles, WDR12 rsl39643391 is occurring in 2 alleles and ADAMTS1 rs2830489 is occurring in 1 allele, or WDR12 rsl39643391 is occurring in 1 allele and ADAMTS1 rs2830489 is occurring in 2 alleles.

[0150] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs based on the subset of subjects carrying two SGV APOE rs429358 alleles (which encodes for the pathogenic APOEE4 isoform; APOEE4 + / + group), is a combination with an odds ratio equal to or higher than 5, i.e. the combinations listed in Table 11A (in lines #1 to and including #56). More in particular, many of these combinations are unique occurrences in Table 11A, except for the combinations of the SGV pairs: INPP5D_rsl0933431 and HS3ST5_rs785129 (occurring in lines #6 and 25 of Table 11A in different combinations of number of risk alleles per SGV);

[0151] TSPAN14_rs6586028 and SLC24A4_rs7401792 (occurring in lines #7 and 10 of Table 11A in different combinations of number of risk alleles per SGV);

[0152] WDR12_rsl39643391 and ANKH_rsll2403360 (occurring in lines #8 and 24 of Table 11A in different combinations of number of risk alleles per SGV);

[0153] SPPL2A_rs8025980 and MAF_rs450674 (occurring in lines #12, 33, 35 and 37 of Table 11A in different combinations of number of risk alleles per SGV); EPDRl_rs6966331 and EPHAl_rsll771145 (occurring in lines #17 and 28 of Table 11A in different combinations of number of risk alleles per SGV);

[0154] EPHAl_rsll771145 and FOXFl_rsl6941239 (occurring in lines #18 and 26 of Table 11A in different combinations of number of risk alleles per SGV);

[0155] HS3ST5_rs785129 and SLC2A4RG_rs6742 (occurring in lines #20 and 21 of Table 11A in different combinations of number of risk alleles per SGV);

[0156] COX7C_rs62374257 and PLEKHAl_rs7908662 (occurring in lines #34 and 47 of Table 11A in different combinations of number of risk alleles per SGV);

[0157] BLNK_rs6584063 and MINDY2_rs602602 (occurring in lines #43 and 53 of Table 11A in different combinations of number of risk alleles per SGV);

[0158] SPDYE3_rs7384878 and PLCG2_rsl2446759 (occurring in lines #44, 54 and 55 of Table 11A in different combinations of number of risk alleles per SGV).

[0159] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs based on the subset of subjects carrying two SGV APOE rs429358 alleles (which encodes for the pathogenic APOEE4 isoform; APOEE4 + / + group), is a combination with an odds ratio equal to or higher than 10, i.e. the combinations listed in Table 11A (in lines #1 to and including #25). More in particular, many of these combinations are unique occurrences in Table 11A, except for the combinations of the SGV pairs: INPP5D_rsl0933431 and HS3ST5_rs785129 (occurring in lines #6 and 25 of Table 11A in different combinations of number of risk alleles per SGV);

[0160] TSPAN14_rs6586028 and SLC24A4_rs7401792 (occurring in lines #7 and 10 of Table 11A in different combinations of number of risk alleles per SGV);

[0161] WDR12_rsl39643391 and ANKH_rsll2403360 (occurring in lines #8 and 24 of Table 11A in different combinations of number of risk alleles per SGV);

[0162] HS3ST5_rs785129 and SLC2A4RG_rs6742 (occurring in lines #20 and 21 of Table 11A in different combinations of number of risk alleles per SGV).

[0163] As such, in particular, a combination of 2 SGVs from the withheld combinations of 2 SGVs based on the subset of subjects carrying two SGV APOE rs429358 alleles (which encodes for the pathogenic APOEE4 isoform; APOEE4 + / + group), is a combination with an odds ratio equal to or higher than 15, i.e. the combinations listed in Table 11A (in lines #1 to and including #7), all of these combinations are unique occurrences in Table 11A.

[0164] Furthermore in particular, the combination of 2 SGVs are JAZFl_rsll60871 and WDR81_rs35048651, or SEC61G_rs76928645 and TPCNl_rs6489896. Furthermore in particular hereto JAZFl_rsll60871 is occurring in one allele and WDR81_rs35048651 in 2 alleles, or SEC61G_rs76928645 and TPCNl_rs6489896 are both occurring in one allele.

[0165] Furthermore in particular, the combination of 2 SGVs is BLNK rs6584063 and SLC24A4 rsl2590654 (occurring in line #9 of Table 9A); TREML2 rs60755019 and BCKDK rs889555 (occurring in lines #7 and 8 of Table 9A in different combinations of number of risk alleles per SGV); TREM2 rs75932628 and ABCA7 rsl2151021 (occurring in line #21 of Table 9A); or MME rsl6824536 and UMAD1 rs6943429 (occurring in lines #11 and 12 of Table IDA in different combinations of number of risk alleles per SGV).

[0166] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs including the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 10, i.e. the combinations listed in Table 5B in lines #1 to and including #39. More in particular, these are combinations of 3 SGVs with one of the 3 SGVs being APOE rs429358, these combinations further including either of the following pairs of SGVs:

[0167] BINI rs6733839 and BLNK rs6584063 (occurring in line #1 of Table 5B);

[0168] TREM2 rsl43332484 and MY015A rs2242595 (occurring in lines #2 and 13 of Table 5B in different combinations of number of risk alleles per SGV);

[0169] UNC5CL rsl0947943 and TREM2 rsl43332484 (occurring in line #3 of Table 5B);

[0170] BINI rs6733839 and SORL1 rsll218343 (occurring in line #4 of Table 5B);

[0171] CD2AP rs7767350 and IGHcluster rsl0131280 (occurring in line #5 of Table 5B);

[0172] JAZF1 rsll60871 and WDR81 rs35048651 (occurring in line #6 of Table 5B);

[0173] FOXF1 rsl6941239 and CASS4 rs6014724 (occurring in lines #7 and 18 of Table 5B in different combinations of number of risk alleles per SGV);

[0174] INPP5D rsl0933431 and TREML2 rs60755019 (occurring in line #8 of Table 5B);

[0175] JAZF1 rsll60871 and SORL1 rs74685827 (occurring in lines #9 and 19 of Table 5B in different combinations of number of risk alleles per SGV);

[0176] CTSB rsl065712 and EED rs3851179 (occurring in lines #10 and 32 of Table 5B in different combinations of number of risk alleles per SGV);

[0177] HS3ST5 rs785129 and SEC61G rs76928645 (occurring in line #11 of Table 5B);

[0178] EPHA1 rsll771145 and CTSB rsl065712 (occurring in lines #12 and 28 of Table 5B in different combinations of number of risk alleles per SGV);

[0179] MY015A rs2242595 and ABI3 rs616338 (occurring in line #14 of Table 5B);

[0180] JAZF1 rsll60871 and IL34 rs4985556 (occurring in line #15 of Table 5B);

[0181] HLAdqal rs6605556 and GRN rs5848 (occurring in line #16 of Table 5B);

[0182] MME rsl6824536 and CTSH rsl2592898 (occurring in line #17 of Table 5B); WDR12 rsl39643391 and S0RL1 rs74685827 (occurring in line #20 of Table 5B);

[0183] ADAM17 rs72777026 and IDUA rs3822030 (occurring in line #21 of Table 5B);

[0184] INPP5D rsl0933431 and TPCN1 rs6489896 (occurring in line #22 of Table 5B);

[0185] PRKD3 rsl7020490 and IDUA rs3822030 (occurring in line #23 of Table 5B);

[0186] INPP5D rsl0933431 and HS3ST5 rs785129 (occurring in line #24 of Table 5B);

[0187] ICA1 rsl0952097 and BCKDK rs889555 (occurring in line #25 of Table 5B);

[0188] SORL1 rsll218343 and CTSH rsl2592898 (occurring in line #26 of Table 5B);

[0189] COX7C rs62374257 and PLEKHA1 rs7908662 (occurring in line #27 of Table 5B);

[0190] WDR12 rsl39643391 and APP rs2154481 (occurring in line #29 of Table 5B);

[0191] USP6NL rs7912495 and WDR81 rs35048651(occurring in line #30 of Table 5B);

[0192] HS3ST5 rs785129 and BCKDK rs889555 (occurring in line #31 of Table 5B);

[0193] TREML2 rs60755019 and KLF16 rsl49080927 (occurring in line #33 of Table 5B);

[0194] UNC5CL rsl0947943 and APP rs2154481 (occurring in line #34 of Table 5B);

[0195] ABI3 rs616338 and RBCK1 rsl358782 (occurring in line #35 of Table 5B);

[0196] USP6NL rs7912495 and CASS4 rs6014724 (occurring in line #36 of Table 5B);

[0197] HLAdqal rs6605556 and ABI3 rs616338 (occurring in line #37 of Table 5B);

[0198] MME rs61762319 and SLC2A4RG rs6742 (occurring in line #38 of Table 5B);

[0199] INPP5D rsl0933431 and MY015A rs2242595 (occurring in line #39 of Table 5B).

[0200] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs including the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 15, i.e. the combinations listed in Table 5B in lines #1 to and including #23. More in particular, these are combinations of 3 SGVs with one of the 3 SGVs being APOE rs429358, these combinations further including either of the following pairs of SGVs:

[0201] BINI rs6733839 and BLNK rs6584063 (occurring in line #1 of Table 5B);

[0202] TREM2 rsl43332484 and MY015A rs2242595 (occurring in lines #2 and 13 of Table 5B in different combinations of number of risk alleles per SGV);

[0203] UNC5CL rsl0947943 and TREM2 rsl43332484 (occurring in line #3 of Table 5B);

[0204] BINI rs6733839 and SORL1 rsll218343 (occurring in line #4 of Table 5B);

[0205] CD2AP rs7767350 and IGHcluster rsl0131280 (occurring in line #5 of Table 5B);

[0206] JAZF1 rsll60871 and WDR81 rs35048651 (occurring in line #6 of Table 5B);

[0207] FOXF1 rsl6941239 and CASS4 rs6014724 (occurring in lines #7 and 18 of Table 5B in different combinations of number of risk alleles per SGV);

[0208] INPP5D rsl0933431 and TREML2 rs60755019 (occurring in line #8 of Table 5B); JAZF1 rsll60871 and S0RL1 rs74685827 (occurring in lines #9 and 19 of Table 5B in different combinations of number of risk alleles per SGV);

[0209] CTSB rsl065712 and EED rs3851179 (occurring in line #10 of Table 5B);

[0210] HS3ST5 rs785129 and SEC61G rs76928645 (occurring in line #11 of Table 5B);

[0211] EPHA1 rsll771145 and CTSB rsl065712 (occurring in lines #12 of Table 5B);

[0212] MY015A rs2242595 and ABI3 rs616338 (occurring in line #14 of Table 5B);

[0213] JAZF1 rsll60871 and IL34 rs4985556 (occurring in line #15 of Table 5B);

[0214] HLAdqal rs6605556 and GRN rs5848 (occurring in line #16 of Table 5B);

[0215] MME rsl6824536 and CTSH rsl2592898 (occurring in line #17 of Table 5B);

[0216] WDR12 rsl39643391 and SORL1 rs74685827 (occurring in line #20 of Table 5B);

[0217] ADAM17 rs72777026 and IDUA rs3822030 (occurring in line #21 of Table 5B);

[0218] INPP5D rsl0933431 and TPCN1 rs6489896 (occurring in line #22 of Table 5B);

[0219] PRKD3 rsl7020490 and IDUA rs3822030 (occurring in line #23 of Table 5B).

[0220] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs including the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 20, i.e. the combinations listed in Table 5B in lines #1 to and including #15. More in particular, these are combinations of 3 SGVs with one of the 3 SGVs being APOE rs429358, these combinations further including either of the following pairs of SGVs:

[0221] BINI rs6733839 and BLNK rs6584063 (occurring in line #1 of Table 5B);

[0222] TREM2 rsl43332484 and MY015A rs2242595 (occurring in lines #2 and 13 of Table 5B in different combinations of number of risk alleles per SGV);

[0223] UNC5CL rsl0947943 and TREM2 rsl43332484 (occurring in line #3 of Table 5B);

[0224] BINI rs6733839 and SORL1 rsll218343 (occurring in line #4 of Table 5B);

[0225] CD2AP rs7767350 and IGHcluster rsl0131280 (occurring in line #5 of Table 5B);

[0226] JAZF1 rsll60871 and WDR81 rs35048651 (occurring in line #6 of Table 5B);

[0227] FOXF1 rsl6941239 and CASS4 rs6014724 (occurring in line #7 of Table 5B);

[0228] INPP5D rsl0933431 and TREML2 rs60755019 (occurring in line #8 of Table 5B);

[0229] JAZF1 rsll60871 and SORL1 rs74685827 (occurring in line #9 of Table 5B);

[0230] CTSB rsl065712 and EED rs3851179 (occurring in line #10 of Table 5B);

[0231] HS3ST5 rs785129 and SEC61G rs76928645 (occurring in line #11 of Table 5B);

[0232] EPHA1 rsll771145 and CTSB rsl065712 (occurring in lines #12 of Table 5B);

[0233] MY015A rs2242595 and ABI3 rs616338 (occurring in line #14 of Table 5B);

[0234] JAZF1 rsll60871 and IL34 rs4985556 (occurring in line #15 of Table 5B). As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs including the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 25, i.e. the combinations listed in Table 5B in lines #1 to and including #9. More in particular, these are combinations of 3 SGVs with one of the 3 SGVs being APOE rs429358, these combinations further including either of the following pairs of SGVs:

[0235] BINI rs6733839 and BLNK rs6584063 (occurring in line #1 of Table 5B);

[0236] TREM2 rsl43332484 and MY015A rs2242595 (occurring in line #2 of Table);

[0237] UNC5CL rsl0947943 and TREM2 rsl43332484 (occurring in line #3 of Table 5B);

[0238] BINI rs6733839 and SORL1 rsll218343 (occurring in line #4 of Table 5B);

[0239] CD2AP rs7767350 and IGHcluster rsl0131280 (occurring in line #5 of Table 5B);

[0240] JAZF1 rsll60871 and WDR81 rs35048651 (occurring in line #6 of Table 5B);

[0241] FOXF1 rsl6941239 and CASS4 rs6014724 (occurring in line #7 of Table 5B);

[0242] INPP5D rsl0933431 and TREML2 rs60755019 (occurring in line #8 of Table 5B);

[0243] JAZF1 rsll60871 and SORL1 rs74685827 (occurring in line #9 of Table 5B).

[0244] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs including the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 30, i.e. the combinations listed in Table 5B in lines #1 to and including #3. More in particular, these are combinations of 3 SGVs with one of the 3 SGVs being APOE rs429358, these combinations further including either of the following pairs of SGVs:

[0245] BINI rs6733839 and BLNK rs6584063 (occurring in line #1 of Table 5B);

[0246] TREM2 rsl43332484 and MY015A rs2242595 (occurring in line #2 of Table 5B);

[0247] UNC5CL rsl0947943 and TREM2 rsl43332484 (occurring in line #3 of Table 5B).

[0248] As such, in particular, a combination of 3 SGVs is a combination of 3 SGVs with one of the 3 SGVs being APOE rs429358, the combination further including SGV UNC5CL rsl0947943 and SGV TREM2 rsl43332484 (occurring in line #3 of Table 5B).

[0249] As such, in particular, a combination of 3 SGVs is a combination of 3 SGVs with one of the 3 SGVs being APOE rs429358, the combination further including SGV TREM2 rsl43332484 and SGV MY015A rs2242595 (occurring in line #2 of Table 5B).

[0250] As such, in particular, a combination of 3 SGVs is a combination of 3 SGVs with one of the 3 SGVs being APOE rs429358, the combination further including SGV BINI rs6733839 and SGV BLNK rs6584063 (occurring in line #1 of Table 5B). As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 5, i.e. the combinations listed in Table 6B in lines #1 to and including #40. More in particular, these are combinations of the following SGV triplets:

[0251] SEC61G rs76928645, EED rs3851179, and ABCA7 rsl2151021 (occurring in lines #1, 3, 5 and 7 of Table6B in different combinations of number of risk alleles per SGV);

[0252] BINI rs6733839, MME rsl6824536, and CLNK rs6846529 (occurring in lines #2 and 4 of Table 6B in different combinations of number of risk alleles per SGV);

[0253] BINI rs6733839, MME rsl6824536, and TSPAN14 rs6586028 (occurring in line #6 of Table 6B);

[0254] BINI rs6733839, RASGEF1C rsll3706587, and TSPAN14 rs6586028 (occurring in lines # 8 and 13 of Table 6B in different combinations of number of risk alleles per SGV);

[0255] BINI rs6733839, MY015A rs2242595, and LILRB2 rs587709 (occurring in lines #9 and 10 of Table 6B in different combinations of number of risk alleles per SGV);

[0256] HLAdqal rs6605556, EED rs3851179, and SNX1 rs3848143 (occurring in lines #11 and 15 of Table 6B in different combinations of number of risk alleles per SGV);

[0257] RASGEF1C rsll3706587, MINDY2 rs602602, and ABCA7 rsl2151021 (occurring in lines #12 and 14 of Table 6B in different combinations of number of risk alleles per SGV);

[0258] PRKD3 rsl7020490, EED rs3851179, and ABCA7 rsl2151021 (occurring in line #16 of Table 6B);

[0259] TSPAN14 rs6586028, MINDY2 rs602602, and ABCA7 rsl2151021 (occurring in lines #17 and 23 of Table 6B in different combinations of number of risk alleles per SGV);

[0260] BINI rs6733839, MME rs61762319, and MINDY2 rs602602 (occurring in lines #18 and 30 of Table 6B in different combinations of number of risk alleles per SGV);

[0261] HLAdqal rs6605556, EED rs3851179, and MINDY2 rs602602(occurring in lines #19 and 21 of Table 6B in different combinations of number of risk alleles per SGV);

[0262] CR1 rs679515, UMAD1 rs6943429, and MINDY2 rs602602 (occurring in lines #20 and 40 of Table 6B in different combinations of number of risk alleles per SGV);

[0263] BINI rs6733839, SORL1 rs74685827, and SLC24A4 rsl2590654 (occurring in lines #22 and 37 of Table 6B in different combinations of number of risk alleles per SGV);

[0264] BINI rs6733839, EPHA1 rsll771145, and APH1B rsll7618017 (occurring in line #24 of Table 6B);

[0265] BINI rs6733839, CLU rsll787077, and PLCG2 rs72824905 (occurring in lines #25 and 33 of Table 6B in different combinations of number of risk alleles per SGV);

[0266] BINI rs6733839, HLAdqal rs6605556, and ABCA7 rsl2151021 (occurring in lines #26 and 27 of Table 6B in different combinations of number of risk alleles per SGV);

[0267] TREM2 rsl43332484, EED rs3851179, and TPCN1 rs6489896 (occurring in line #28 of Table 6B); BINI rs6733839, SPI1 rsl0437655, and IGHcluster rsl0131280 (occurring in lines #29 and 38 of Table 6B in different combinations of number of risk alleles per SGV);

[0268] ADAM17 rs72777026, BINI rs6733839, and UMAD1 rs6943429 (occurring in line #31 of Table 6B);

[0269] BINI rs6733839, CTSH rsl2592898, and ABCA7 rsl2151021 (occurring in lines #32 and 39 of Table 6B in different combinations of number of risk alleles per SGV);

[0270] ADAM17 rs72777026, BINI rs6733839, and SLC24A4 rsl2590654 (occurring in line #34 of Table 6B); INPP5D rsl0933431, EED rs3851179, and ABCA7 rsl2151021 (occurring in line #35 of Table 6B);

[0271] BINI rs6733839, RASGEF1C rsll3706587, and CD2AP rs7767350 (occurring in line #36 of Table 6B).

[0272] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 10, i.e. the combinations listed in Table 6B in lines #1 to and including #16. More in particular, these are combinations of the following SGV triplets:

[0273] SEC61G rs76928645, EED rs3851179, and ABCA7 rsl2151021 (occurring in lines #1, 3, 5 and 7 of Table6B in different combinations of number of risk alleles per SGV);

[0274] BINI rs6733839, MME rsl6824536, and CLNK rs6846529 (occurring in lines #2 and 4 of Table 6B in different combinations of number of risk alleles per SGV);

[0275] BINI rs6733839, MME rsl6824536, and TSPAN14 rs6586028 (occurring in line #6 of Table 6B);

[0276] BINI rs6733839, RASGEF1C rsll3706587, and TSPAN14 rs6586028 (occurring in lines # 8 and 13 of Table 6B in different combinations of number of risk alleles per SGV);

[0277] BINI rs6733839, MY015A rs2242595, and LILRB2 rs587709 (occurring in lines #9 and 10 of Table 6B in different combinations of number of risk alleles per SGV);

[0278] HLAdqal rs6605556, EED rs3851179, and SNX1 rs3848143 (occurring in lines #11 and 15 of Table 6B in different combinations of number of risk alleles per SGV);

[0279] RASGEF1C rsll3706587, MINDY2 rs602602, and ABCA7 rsl2151021 (occurring in lines #12 and 14 of Table 6B in different combinations of number of risk alleles per SGV);

[0280] PRKD3 rsl7020490, EED rs3851179, and ABCA7 rsl2151021 (occurring in line #16 of Table 6B).

[0281] Selected combinations of 3 or 4 SGVs identified without prior stratification of the training dataset according to number of APOEE4 alleles / according to APOEE4 background are listed hereinafter.

[0282] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 15, i.e. the combinations listed in Table 6B in lines #1 to and including #15. More in particular, these are combinations of the following SGV triplets: SEC61G rs76928645, EED rs3851179, and ABCA7 rsl2151021 (occurring in lines #1, 3, 5 and 7 of Table6B in different combinations of number of risk alleles per SGV);

[0283] BINI rs6733839, MME rsl6824536, and CLNK rs6846529 (occurring in lines #2 and 4 of Table 6B in different combinations of number of risk alleles per SGV);

[0284] BINI rs6733839, MME rsl6824536, and TSPAN14 rs6586028 (occurring in line #6 of Table 6B);

[0285] BINI rs6733839, RASGEF1C rsll3706587, and TSPAN14 rs6586028 (occurring in lines # 8 and 13 of Table 6B in different combinations of number of risk alleles per SGV);

[0286] BINI rs6733839, MY015A rs2242595, and LILRB2 rs587709 (occurring in lines #9 and 10 of Table 6B in different combinations of number of risk alleles per SGV);

[0287] HLAdqal rs6605556, EED rs3851179, and SNX1 rs3848143 (occurring in lines #11 and 15 of Table 6B in different combinations of number of risk alleles per SGV);

[0288] RASGEF1C rsll3706587, MINDY2 rs602602, and ABCA7 rsl2151021 (occurring in lines #12 and 14 of Table 6B in different combinations of number of risk alleles per SGV).

[0289] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 20, i.e. the combinations listed in Table 6B in lines #1 to and including #12. More in particular, these are combinations of the following SGV triplets:

[0290] SEC61G rs76928645, EED rs3851179, and ABCA7 rsl2151021 (occurring in lines #1, 3, 5 and 7 of Table7B in different combinations of number of risk alleles per SGV);

[0291] BINI rs6733839, MME rsl6824536, and CLNK rs6846529 (occurring in lines #2 and 4 of Table 7B in different combinations of number of risk alleles per SGV);

[0292] BINI rs6733839, MME rsl6824536, and TSPAN14 rs6586028 (occurring in line #6 of Table 6B);

[0293] BINI rs6733839, RASGEF1C rsll3706587, and TSPAN14 rs6586028 (occurring in line #8 of Table 6B) ;

[0294] BINI rs6733839, MY015A rs2242595, and LILRB2 rs587709 (occurring in lines #9 and 10 of Table 7B in different combinations of number of risk alleles per SGV);

[0295] HLAdqal rs6605556, EED rs3851179, and SNX1 rs3848143 (occurring in line #11 of Table 6B);

[0296] RASGEF1C rsll3706587, MINDY2 rs602602, and ABCA7 rsl2151021 (occurring in line #12 of Table 6B).

[0297] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 25, i.e. the combinations listed in Table 6B in lines #1 to and including #8. More in particular, these are combinations of the following SGV triplets:

[0298] SEC61G rs76928645, EED rs3851179, and ABCA7 rsl2151021 (occurring in lines #1, 3, 5 and 7 of Table7B in different combinations of number of risk alleles per SGV); BINI rs6733839, MME rsl6824536, and CLNK rs6846529 (occurring in lines #2 and 4 of Table 7B in different combinations of number of risk alleles per SGV);

[0299] BINI rs6733839, MME rsl6824536, and TSPAN14 rs6586028 (occurring in line #6 of Table 6B); BINI rs6733839, RASGEF1C rsll3706587, and TSPAN14 rs6586028(occurring in line #8 of Table 6B).

[0300] As such, in particular, a combination of 3 SGVs is a combination of the SGV triplets BINI rs6733839, MME rsl6824536, and CLNK rs6846529 (occurring in lines #2 and 4 of Table 7B).

[0301] As such, in particular, a combination of 3 SGVs is a combination of the SGV triplets SEC61G rs76928645, EED rs3851179, and ABCA7 rsl2151021 (occurring in lines #1, 3, 5 and 7 of Table7B).

[0302] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs including the SGV APOE rs429358 is a combination comprising the SGV APOE rs429358 and a further SGV with a frequency of occurrence (Table 5C) of at least 10 out of the 304 withheld combinations. More in particular, such combinations of 3 SGVs are comprising the SGV APOE rs429358 and at least one of the SGVs chosen from: HS3ST5 rs785129, INPP5D rsl0933431, BINI rs6733839, LILRB2 rs587709, UMAD1 rs6943429, CR1 rs679515, PRKD3 rsl7020490, USP6NL rs7912495, ADAM17 rs72777026, CD2AP rs7767350, SPDYE3 rs7384878, TNIP1 rs871269, PLEKHA1 rs7908662, CLU rsll787077, COX7C rs62374257, ADAMTS1 rs2830489, CLNK rs6846529, EED rs3851179, MINDY2 rs602602, and MY015A rs2242595. In one embodiment thereto, the SGV APOE rs429358 and the chosen SGV are combined with a third SGV of Table 1 wherein the third SGV is different from the SGV APOE rs429358 and different from the chosen SGV.

[0303] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs including the SGV APOE rs429358 is a combination comprising the SGV APOE rs429358 and a further SGV with a frequency of occurrence (Table 5C) of at least 11 out of the 304 withheld combinations. More in particular, such combinations of 3 SGVs are comprising the SGV APOE rs429358 and at least one of the SGVs chosen from: HS3ST5 rs785129, INPP5D rsl0933431, BINI rs6733839, LILRB2 rs587709, UMAD1 rs6943429, CR1 rs679515, PRKD3 rsl7020490, USP6NL rs7912495, ADAM17 rs72777026, CD2AP rs7767350, SPDYE3 rs7384878, TNIP1 rs871269, PLEKHA1 rs7908662, CLU rsll787077, and COX7C rs62374257. In one embodiment thereto, the SGV APOE rs429358 and the chosen SGV are combined with a third SGV of Table 1 wherein the third SGV is different from the SGV APOE rs429358 and different from the chosen SGV. As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs including the SGV APOE rs429358 is a combination comprising the SGV APOE rs429358 and a further SGV with a frequency of occurrence (Table 5C) of at least 12 out of the 304 withheld combinations. More in particular, such combinations of 3 SGVs are comprising the SGV APOE rs429358 and at least one of the SGVs chosen from: HS3ST5 rs785129, INPP5D rsl0933431, BINI rs6733839, LILRB2 rs587709, UMAD1 rs6943429, CR1 rs679515, PRKD3 rsl7020490, USP6NL rs7912495, ADAM17 rs72777026, CD2AP rs7767350, SPDYE3 rs7384878, TNIP1 rs871269, and PLEKHA1 rs7908662. In one embodiment thereto, the SGV APOE rs429358 and the chosen SGV are combined with a third SGV of Table 1 wherein the third SGV is different from the SGV APOE rs429358 and different from the chosen SGV.

[0304] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs including the SGV APOE rs429358 is a combination comprising the SGV APOE rs429358 and a further SGV with a frequency of occurrence (Table 5C) of at least 13 out of the 304 withheld combinations. More in particular, such combinations of 3 SGVs are comprising the SGV APOE rs429358 and at least one of the SGVs chosen from: HS3ST5 rs785129, INPP5D rsl0933431, BINI rs6733839, LILRB2 rs587709, UMAD1 rs6943429, CR1 rs679515, PRKD3 rsl7020490, USP6NL rs7912495, ADAM17 rs72777026, CD2AP rs7767350, SPDYE3 rs7384878, and TNIP1 rs871269. In one embodiment thereto, the SGV APOE rs429358 and the chosen SGV are combined with a third SGV of Table 1 wherein the third SGV is different from the SGV APOE rs429358 and different from the chosen SGV.

[0305] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs including the SGV APOE rs429358 is a combination comprising the SGV APOE rs429358 and a further SGV with a frequency of occurrence (Table 5C) of at least 14 out of the 304 withheld combinations. More in particular, such combinations of 3 SGVs are comprising the SGV APOE rs429358 and at least one of the SGVs chosen from: HS3ST5 rs785129, INPP5D rsl0933431, BINI rs6733839, LILRB2 rs587709, UMAD1 rs6943429, CR1 rs679515, PRKD3 rsl7020490, and USP6NL rs7912495. In one embodiment thereto, the SGV APOE rs429358 and the chosen SGV are combined with a third SGV of Table 1 wherein the third SGV is different from the SGV APOE rs429358 and different from the chosen SGV.

[0306] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs including the SGV APOE rs429358 is a combination comprising the SGV APOE rs429358 and a further SGV with a frequency of occurrence (Table 5C) of at least 15 out of the 304 withheld combinations. More in particular, such combinations of 3 SGVs are comprising the SGV APOE rs429358 and at least one of the SGVs chosen from: HS3ST5 rs785129, INPP5D rsl0933431, BINI rs6733839, LILRB2 rs587709, and UMAD1 rs6943429. In one embodiment thereto, the SGV APOE rs429358 and the chosen SGV are combined with a third SGV of Table 1 wherein the third SGV is different from the SGV APOE rs429358 and different from the chosen SGV.

[0307] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs including the SGV APOE rs429358 is a combination comprising the SGV APOE rs429358 and the SGV LILRB2 rs587709. In one embodiment thereto, the SGV APOE rs429358 and the SGV LILRB2 rs587709 are combined with a third SGV of Table 1 wherein the third SGV is different from the SGV APOE rs429358 and the SGV LILRB2 rs587709.

[0308] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs including the SGV APOE rs429358 is a combination comprising the SGV APOE rs429358 and the SGV BINI rs6733839. In one embodiment thereto, the SGV APOE rs429358 and the SGV BINI rs6733839 are combined with a third SGV of Table 1 wherein the third SGV is different from the SGV APOE rs429358 and the SGV BINI rs6733839.

[0309] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs including the SGV APOE rs429358 is a combination comprising the SGV APOE rs429358 and the SGV INPP5D rsl0933431. In one embodiment thereto, the SGV APOE rs429358 and the SGV INPP5D rsl0933431 are combined with a third SGV of Table 1 wherein the third SGV is different from the SGV APOE rs429358 and the SGV INPP5D rsl0933431.

[0310] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs including the SGV APOE rs429358 is a combination comprising the SGV APOE rs429358 and the SGV HS3ST5 rs785129. In one embodiment thereto, the SGV APOE rs429358 and the SGV HS3ST5 rs785129 are combined with a third SGV of Table 1 wherein the third SGV is different from the SGV APOE rs429358 and the SGV HS3ST5 rs785129.

[0311] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs excluding the SGV APOE rs429358 is a combination comprising one SGV with a frequency of occurrence (Table 6C) of at least 6 out of the 140 withheld combinations. More in particular, such combinations of 3 SGVs are comprising at least one of the SGVs chosen from: BINI rs6733839, EED rs3851179, ABCA7 rsl2151021, CLNK rs6846529, MINDY2 rs602602, SLC24A4 rsl2590654, LILRB2 rs587709, SPI1 rsl0437655, EPHA1 rsll771145, CLU rsll787077, HLAdqal rs6605556, IGHcluster rsl0131280, MAF rs450674, PLEKHA1 rs7908662, RASGEF1C rsll3706587, TSPAN14 rs6586028, and UMAD1 rs6943429. In one embodiment thereto, the chosen SGV is combined with a second SGV and a third SGV of Table 1 wherein the second SGV and the third SGV each are different from the chosen SGV.

[0312] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs excluding the SGV APOE rs429358 is a combination comprising one SGV with a frequency of occurrence (Table 6C) of at least 7 out of the 140 withheld combinations. More in particular, such combinations of 3 SGVs are comprising at least one of the SGVs chosen from: BINI rs6733839, EED rs3851179, ABCA7 rsl2151021, CLNK rs6846529, MINDY2 rs602602, SLC24A4 rsl2590654, LILRB2 rs587709, SPI1 rsl0437655, EPHA1 rsll771145, CLU rsll787077, HLAdqal rs6605556, IGHcluster rsl0131280, and MAF rs450674. In one embodiment thereto, the chosen SGV is combined with a second SGV and a third SGV of Table 1 wherein the second SGV and the third SGV each are different from the chosen SGV.

[0313] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs excluding the SGV APOE rs429358 is a combination comprising one SGV with a frequency of occurrence (Table 6C) of at least 8 out of the 140 withheld combinations. More in particular, such combinations of 3 SGVs are comprising at least one of the SGVs chosen from: BINI rs6733839, EED rs3851179, ABCA7 rsl2151021, CLNK rs6846529, MINDY2 rs602602, SLC24A4 rsl2590654, LILRB2 rs587709, SPI1 rsl0437655, EPHA1 rsll771145, CLU rsll787077, and HLAdqal rs6605556. In one embodiment thereto, the chosen SGV is combined with a second SGV and a third SGV of Table 1 wherein the second SGV and the third SGV each are different from the chosen SGV.

[0314] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs excluding the SGV APOE rs429358 is a combination comprising one SGV with a frequency of occurrence (Table 6C) of at least 11 out of the 140 withheld combinations. More in particular, such combinations of 3 SGVs are comprising at least one of the SGVs chosen from: BINI rs6733839, EED rs3851179, ABCA7 rsl2151021, CLNK rs6846529, MINDY2 rs602602, SLC24A4 rsl2590654, LILRB2 rs587709, SPI1 rsl0437655, and EPHA1 rsll771145. In one embodiment thereto, the chosen SGV is combined with a second SGV and a third SGV of Table 1 wherein the second SGV and the third SGV each are different from the chosen SGV.

[0315] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs excluding the SGV APOE rs429358 is a combination comprising one SGV with a frequency of occurrence (Table 6C) of at least 14 out of the 140 withheld combinations. More in particular, such combinations of 3 SGVs are comprising at least one of the SGVs chosen from: BINI rs6733839, EED rs3851179, ABCA7 rsl2151021, CLNK rs6846529, MINDY2 rs602602, SLC24A4 rsl2590654, LILRB2 rs587709, and SPI1 rsl0437655. In one embodiment thereto, the chosen SGV is combined with a second SGV and a third SGV of Table 1 wherein the second SGV and the third SGV each are different from the chosen SGV.

[0316] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs excluding the SGV APOE rs429358 is a combination comprising one SGV with a frequency of occurrence (Table 6C) of at least 16 out of the 140 withheld combinations. More in particular, such combinations of 3 SGVs are comprising at least one of the SGVs chosen from: BINI rs6733839, EED rs3851179, ABCA7 rsl2151021, CLNK rs6846529, MINDY2 rs602602, and SLC24A4 rsl2590654. In one embodiment thereto, the chosen SGV is combined with a second SGV and a third SGV of Table 1 wherein the second SGV and the third SGV each are different from the chosen SGV.

[0317] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs excluding the SGV APOE rs429358 is a combination of 3 SGVs comprising the SGV MINDY2 rs602602 or the SGV CLNK rs6846529. In one embodiment thereto, the chosen SGV is combined with a second SGV and a third SGV of Table 1 wherein the second SGV and the third SGV each are different from the chosen SGV.

[0318] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs excluding the SGV APOE rs429358 is a combination of 3 SGVs comprising the SGV ABCA7 rsl2151021. In one embodiment thereto, the SGV ABCA7 rsl2151021 is combined with a second SGV and a third SGV of Table 1 wherein the second SGV and the third SGV each are different from the SGV ABCA7 rsl2151021.

[0319] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs excluding the SGV APOE rs429358 is a combination of 3 SGVs comprising the SGV EED rs3851179. In one embodiment thereto, the SGV EED rs3851179 is combined with a second SGV and a third SGV of Table 1 wherein the second SGV and the third SGV each are different from the SGV EED rs3851179.

[0320] As such, in particular, a combination of 3 SGVs from the withheld combinations of 3 SGVs excluding the SGV APOE rs429358 is a combination of 3 SGVs comprising the SGV BINI rs6733839. In one embodiment thereto, the SGV BINI rs6733839 is combined with a second SGV and a third SGV of Table 1 wherein the second SGV and the third SGV each are different from the SGV BINI rs6733839.

[0321] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs including the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 5, i.e. the combinations listed in Table 7B in lines #1 to and including #22. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being APOE rs429358, these combinations further including either of the following triplets of SGVs:

[0322] BINI rs6733839, TNIP1 rs871269, and TPCN1 rs6489896 (occurring in line #1 of Table 7B);

[0323] MME rs61762319, SHARPIN rs34173062, and MY015A rs2242595 (occurring in lines #2 and 3 of Table 7B in different combinations of number of risk alleles per SGV);

[0324] RHOH rs2245466, RASGEF1C rsll3706587, and SHARPIN rs34173062 (occurring in lines #4 and 12 of Table 7B in different combinations of number of risk alleles per SGV);

[0325] BINI rs6733839, SHARPIN rs34173062, and IGHcluster rs7157106 (occurring in line #5 of Table 7B);

[0326] SHARPIN rs34173062, SCIMP rs7225151, and MY015A rs2242595 (occurring in lines #6 and 7 of Table 7B in different combinations of number of risk alleles per SGV);

[0327] SHARPIN rs34173062, PLEKHA1 rs7908662, and IGHcluster rs7157106 (occurring in lines #8 and 10 of Table 7B in different combinations of number of risk alleles per SGV);

[0328] EPDR1 rs6966331, SHARPIN rs34173062, and SNX1 rs3848143 (occurring in line #9 of Table 7B);

[0329] RASGEF1C rsll3706587, SHARPIN rs34173062, and SLC24A4 rs7401792 (occurring in line #11 of Table 7B);

[0330] SHARPIN rs34173062, MS4A4A rsl582763, and SNX1 rs3848143 (occurring in line #13 of Table 7B); RHOH rs2245466, SHARPIN rs34173062, and SNX1 rs3848143 (occurring in line #14 of Table 7B); PTK2B rs73223431, SHARPIN rs34173062, and SCIMP rs7225151 (occurring in line #15 of Table 7B); CLNK rs6846529, RASGEF1C rsll3706587, and LILRB2 rs587709 (occurring in line #16 of Table 7B); COX7C rs62374257, PTK2B rs73223431, and SHARPIN rs34173062 (occurring in line #17 of Table 7B);

[0331] SHARPIN rs34173062, SLC24A4 rs7401792, and DOC2A rsll40239 (occurring in line #18 of Table 7B); COX7C rs62374257, SHARPIN rs34173062, and GRN rs5848 (occurring in line #19 of Table 7B);

[0332] RASGEF1C rsll3706587, SHARPIN rs34173062, and LILRB2 rs587709 (occurring in line #20 of Table 7B); SHARPIN rs34173062, SNX1 rs3848143, and TSPOAP1 rs2526377 (occurring in line #21 of Table 7B); ANK3 rs7068231, TSPOAP1 rs2526377, and APOE rs7412 (occurring in line #22 of Table 7B).

[0333] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs including the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 10, i.e. the combinations listed in Table 7B in lines #1 to and including #16. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being APOE rs429358, these combinations further including either of the following triplets of SGVs:

[0334] BINI rs6733839, TNIP1 rs871269, and TPCN1 rs6489896 (occurring in line #1 of Table 7B);

[0335] MME rs61762319, SHARPIN rs34173062, and MY015A rs2242595 (occurring in lines #2 and 3 of Table 7B in different combinations of number of risk alleles per SGV); RHOH rs2245466, RASGEF1C rsll3706587, and SHARPIN rs34173062 (occurring in lines #4 and 12 of Table 7B in different combinations of number of risk alleles per SGV);

[0336] BINI rs6733839, SHARPIN rs34173062, and IGHcluster rs7157106 (occurring in line #5 of Table 7B);

[0337] SHARPIN rs34173062, SCIMP rs7225151, and MY015A rs2242595 (occurring in lines #6 and 7 of Table 7B in different combinations of number of risk alleles per SGV);

[0338] SHARPIN rs34173062, PLEKHA1 rs7908662, and IGHcluster rs7157106 (occurring in lines #8 and 10 of Table 7B in different combinations of number of risk alleles per SGV);

[0339] EPDR1 rs6966331, SHARPIN rs34173062, and SNX1 rs3848143 (occurring in line #9 of Table 7B);

[0340] RASGEF1C rsll3706587, SHARPIN rs34173062, and SLC24A4 rs7401792 (occurring in line #11 of Table 7B);

[0341] SHARPIN rs34173062, MS4A4A rsl582763, and SNX1 rs3848143 (occurring in line #13 of Table 7B); RHOH rs2245466, SHARPIN rs34173062, and SNX1 rs3848143 (occurring in line #14 of Table 7B); PTK2B rs73223431, SHARPIN rs34173062, and SCIMP rs7225151 (occurring in line #15 of Table 7B); CLNK rs6846529, RASGEF1C rsll3706587, and LILRB2 rs587709 (occurring in line #16 of Table 7B).

[0342] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs including the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 20, i.e. the combinations listed in Table 7B in lines #1 to and including #14. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being APOE rs429358, these combinations further including either of the following triplets of SGVs:

[0343] BINI rs6733839, TNIP1 rs871269, and TPCN1 rs6489896 (occurring in line #1 of Table 7B);

[0344] MME rs61762319, SHARPIN rs34173062, and MY015A rs2242595 (occurring in lines #2 and 3 of Table 7B in different combinations of number of risk alleles per SGV);

[0345] RHOH rs2245466, RASGEF1C rsll3706587, and SHARPIN rs34173062 (occurring in lines #4 and 12 of Table 7B in different combinations of number of risk alleles per SGV);

[0346] BINI rs6733839, SHARPIN rs34173062, and IGHcluster rs7157106 (occurring in line #5 of Table 7B);

[0347] SHARPIN rs34173062, SCIMP rs7225151, and MY015A rs2242595 (occurring in lines #6 and 7 of Table 7B in different combinations of number of risk alleles per SGV);

[0348] SHARPIN rs34173062, PLEKHA1 rs7908662, and IGHcluster rs7157106 (occurring in lines #8 and 10 of Table 7B in different combinations of number of risk alleles per SGV);

[0349] EPDR1 rs6966331, SHARPIN rs34173062, and SNX1 rs3848143 (occurring in line #9 of Table 7B);

[0350] RASGEF1C rsll3706587, SHARPIN rs34173062, and SLC24A4 rs7401792 (occurring in line #11 of Table 7B);

[0351] SHARPIN rs34173062, MS4A4A rsl582763, and SNX1 rs3848143 (occurring in line #13 of Table 7B); RHOH rs2245466, SHARPIN rs34173062, and SNX1 rs3848143 (occurring in line #14 of Table 7B). As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs including the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 25, i.e. the combinations listed in Table 7B in lines #1 to and including #12. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being APOE rs429358, these combinations further including either of the following triplets of SGVs:

[0352] BINI rs6733839, TNIP1 rs871269, and TPCN1 rs6489896 (occurring in line #1 of Table 7B);

[0353] MME rs61762319, SHARPIN rs34173062, and MY015A rs2242595 (occurring in lines #2 and 3 of Table 7B in different combinations of number of risk alleles per SGV);

[0354] RHOH rs2245466, RASGEF1C rsll3706587, and SHARPIN rs34173062 (occurring in lines #4 and 12 of Table 7B in different combinations of number of risk alleles per SGV);

[0355] BINI rs6733839, SHARPIN rs34173062, and IGHcluster rs7157106 (occurring in line #5 of Table 7B);

[0356] SHARPIN rs34173062, SCIMP rs7225151, and MY015A rs2242595 (occurring in lines #6 and 7 of Table 7B in different combinations of number of risk alleles per SGV);

[0357] SHARPIN rs34173062, PLEKHA1 rs7908662, and IGHcluster rs7157106 (occurring in lines #8 and 10 of Table 7B in different combinations of number of risk alleles per SGV);

[0358] EPDR1 rs6966331, SHARPIN rs34173062, and SNX1 rs3848143 (occurring in line #9 of Table 7B);

[0359] RASGEF1C rsll3706587, SHARPIN rs34173062, and SLC24A4 rs7401792 (occurring in line #11 of Table 7B).

[0360] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs including the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 30, i.e. the combinations listed in Table 7B in lines #1 to and including #10. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being APOE rs429358, these combinations further including either of the following triplets of SGVs:

[0361] BINI rs6733839, TNIP1 rs871269, and TPCN1 rs6489896 (occurring in line #1 of Table 7B);

[0362] MME rs61762319, SHARPIN rs34173062, and MY015A rs2242595 (occurring in lines #2 and 3 of Table 7B in different combinations of number of risk alleles per SGV);

[0363] RHOH rs2245466, RASGEF1C rsll3706587, and SHARPIN rs34173062 (occurring in line #4 of Table 7B); BINI rs6733839, SHARPIN rs34173062, and IGHcluster rs7157106 (occurring in line #5 of Table 7B);

[0364] SHARPIN rs34173062, SCIMP rs7225151, and MY015A rs2242595 (occurring in lines #6 and 7 of Table 7B in different combinations of number of risk alleles per SGV);

[0365] SHARPIN rs34173062, PLEKHA1 rs7908662, and IGHcluster rs7157106 (occurring in lines #8 and 10 of Table 7B in different combinations of number of risk alleles per SGV);

[0366] EPDR1 rs6966331, SHARPIN rs34173062, and SNX1 rs3848143 (occurring in line #9 of Table 7B). As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs including the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 40, i.e. the combinations listed in Table 7B in lines #1 to and including #9. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being APOE rs429358, these combinations further including either of the following triplets of SGVs:

[0367] BINI rs6733839, TNIP1 rs871269, and TPCN1 rs6489896 (occurring in line #1 of Table 7B);

[0368] MME rs61762319, SHARPIN rs34173062, and MY015A rs2242595 (occurring in lines #2 and 3 of Table 7B in different combinations of number of risk alleles per SGV);

[0369] RHOH rs2245466, RASGEF1C rsll3706587, and SHARPIN rs34173062 (occurring in line #4 of Table 7B); BINI rs6733839, SHARPIN rs34173062, and IGHcluster rs7157106 (occurring in line #5 of Table 7B);

[0370] SHARPIN rs34173062, SCIMP rs7225151, and MY015A rs2242595 (occurring in lines #6 and 7 of Table 7B in different combinations of number of risk alleles per SGV);

[0371] SHARPIN rs34173062, PLEKHA1 rs7908662, and IGHcluster rs7157106 (occurring in line #8 of Table 7B); EPDR1 rs6966331, SHARPIN rs34173062, and SNX1 rs3848143 (occurring in line #9 of Table 7B).

[0372] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs including the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 50, i.e. the combinations listed in Table 7B in lines #1 to and including #7. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being APOE rs429358, these combinations further including either of the following triplets of SGVs:

[0373] BINI rs6733839, TNIP1 rs871269, and TPCN1 rs6489896 (occurring in line #1 of Table 7B);

[0374] MME rs61762319, SHARPIN rs34173062, and MY015A rs2242595 (occurring in lines #2 and 3 of Table 7B in different combinations of number of risk alleles per SGV);

[0375] RHOH rs2245466, RASGEF1C rsll3706587, and SHARPIN rs34173062 (occurring in line #4 of Table 7B); BINI rs6733839, SHARPIN rs34173062, and IGHcluster rs7157106 (occurring in line #5 of Table 7B);

[0376] SHARPIN rs34173062, SCIMP rs7225151, and MY015A rs2242595 (occurring in lines #6 and 7 of Table 7B in different combinations of number of risk alleles per SGV).

[0377] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs including the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 60, i.e. the combinations listed in Table 7B in lines #1 to and including #6. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being APOE rs429358, these combinations further including either of the following triplets of SGVs:

[0378] BINI rs6733839, TNIP1 rs871269, and TPCN1 rs6489896 (occurring in line #1 of Table 7B); MME rs61762319, SHARPIN rs34173062, and MY015A rs2242595 (occurring in lines #2 and 3 of Table 7B in different combinations of number of risk alleles per SGV);

[0379] RHOH rs2245466, RASGEF1C rsll3706587, and SHARPIN rs34173062 (occurring in line #4 of Table 7B); BINI rs6733839, SHARPIN rs34173062, and IGHcluster rs7157106 (occurring in line #5 of Table 7B); SHARPIN rs34173062, SCIMP rs7225151, and MY015A rs2242595 (occurring in line #6 of Table 7B).

[0380] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs including the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 100, i.e. the combinations listed in Table 7B in lines #1 to and including #5. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being APOE rs429358, these combinations further including either of the following triplets of SGVs:

[0381] BINI rs6733839, TNIP1 rs871269, and TPCN1 rs6489896 (occurring in line #1 of Table 7B);

[0382] MME rs61762319, SHARPIN rs34173062, and MY015A rs2242595 (occurring in lines #2 and 3 of Table 7B in different combinations of number of risk alleles per SGV);

[0383] RHOH rs2245466, RASGEF1C rsll3706587, and SHARPIN rs34173062 (occurring in line #4 of Table 7B); BINI rs6733839, SHARPIN rs34173062, and IGHcluster rs7157106 (occurring in line #5 of Table 7B).

[0384] As such, in particular, a combination of 4 SGVs is a combination of the 4 SGVs APOE rs429358, RHOH rs2245466, RASGEF1C rsll3706587, and SHARPIN rs34173062 (occurring in line #4 of Table 7B).

[0385] As such, in particular, a combination of 4 SGVs is a combination of the 4 SGVs APOE rs429358, MME rs61762319, SHARPIN rs34173062, and MY015A rs2242595 (occurring in lines #2 and 3 of Table 7B).

[0386] As such, in particular, a combination of 4 SGVs is a combination of the 4 SGVs APOE rs429358, BINI rs6733839, TNIP1 rs871269, and TPCN1 rs6489896 (occurring in line #1 of Table 7B).

[0387] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 20, i.e. the combinations listed in Table 8B in lines #1 to and including #29. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being BINI rs6733839, these combinations further include either of the following triplets of SGVs:

[0388] RASGEF1C rsll3706587, HLAdqal rs6605556, and EED rs3851179 (occurring in line #1 of Table 8B);

[0389] HLAdqal rs6605556, EED rs3851179, and MAF rs450674 (occurring in lines #2, 5, and 7 of Table 8B in different combinations of number of risk alleles per SGV);

[0390] HLAdqal rs6605556, MS4A4A rsl582763, and ABCA7 rsl2151021 (occurring in line #3 of Table 8B); HLAdqal rs6605556, MINDY2 rs602602, and ADAMTS1 rs2830489 (occurring in line #4 of Table 8B);

[0391] HLAdqal rs6605556, UMAD1 rs6943429, and MINDY2 rs602602 (occurring in lines #6, 10, 13 and 18 of Table 8B in different combinations of number of risk alleles per SGV);

[0392] HLAdqal rs6605556, CLU rsll787077, and EED rs3851179 (occurring in line #8 of Table 8B);

[0393] CR1 rs679515, EED rs3851179, and TPCN1 rs6489896 (occurring in line #9 of Table 8B);

[0394] SPI1 rsl0437655, EED rs3851179, and RBCK1 rsl358782 (occurring in lines #11 and 19 of Table 8B in different combinations of number of risk alleles per SGV);

[0395] HLAdqal rs6605556, EED rs3851179, and SNX1 rs3848143 (occurring in lines #12 and 15 of Table 8B in different combinations of number of risk alleles per SGV);

[0396] HS3ST5 rs785129, EPHA1 rsll771145, and EED rs3851179 (occurring in line #14 of Table 8B);

[0397] CLNK rs6846529, CLU rsll787077, and ABCA7 rsl2151021 (occurring in lines #16 and 29 of Table 8B in different combinations of number of risk alleles per SGV);

[0398] HLAdqal rs6605556, EED rs3851179, and DOC2A rsll40239 (occurring in line #17 of Table 8B);

[0399] HLAdqal rs6605556, CLU rsll787077, and KLF16 rsl49080927 (occurring in line #20 of Table 8B);

[0400] CLNK rs6846529, EPHA1 rsll771145, and CLU rsll787077 (occurring in lines #21, 22 and 25 of Table 8B in different combinations of number of risk alleles per SGV);

[0401] COX7C rs62374257, EED rs3851179, and TPCN1 rs6489896 (occurring in line #23 of Table 8B);

[0402] CLU rsll787077, EED rs3851179, and WDR81 rs35048651 (occurring in line #24 of Table 8B);

[0403] HLAdqal rs6605556, PTK2B rs73223431, and EED rs3851179 (occurring in line #26 of Table 8B); HLAdqal rs6605556, SLC24A4 rs7401792, and SNX1 rs3848143 (occurring in line #27 of Table 8B); EPHA1 rsll771145, EED rs3851179, and SLC2A4RG rs6742 (occurring in line #28 of Table 8B).

[0404] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 25, i.e. the combinations listed in Table 8B in lines #1 to and including #24. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being BINI rs6733839, these combinations further including either of the following triplets of SGVs:

[0405] RASGEF1C rsll3706587, HLAdqal rs6605556, and EED rs3851179 (occurring in line #1 of Table 8B);

[0406] HLAdqal rs6605556, EED rs3851179, and MAF rs450674 (occurring in lines #2, 5, and 7 of Table 8B in different combinations of number of risk alleles per SGV);

[0407] HLAdqal rs6605556, MS4A4A rsl582763, and ABCA7 rsl2151021 (occurring in line #3 of Table 8B); HLAdqal rs6605556, MINDY2 rs602602, and ADAMTS1 rs2830489 (occurring in line #4 of Table 8B); HLAdqal rs6605556, UMAD1 rs6943429, and MINDY2 rs602602 (occurring in lines #6, 10, 13 and 18 of Table 8B in different combinations of number of risk alleles per SGV);

[0408] HLAdqal rs6605556, CLU rsll787077, and EED rs3851179 (occurring in line #8 of Table 8B); CR1 rs679515, EED rs3851179, and TPCN1 rs6489896 (occurring in line #9 of Table 8B);

[0409] SPI1 rsl0437655, EED rs3851179, and RBCK1 rsl358782 (occurring in lines #11 and 19 of Table 8B in different combinations of number of risk alleles per SGV);

[0410] HLAdqal rs6605556, EED rs3851179, and SNX1 rs3848143 (occurring in lines #12 and 15 of Table 8B in different combinations of number of risk alleles per SGV);

[0411] HS3ST5 rs785129, EPHA1 rsll771145, and EED rs3851179 (occurring in line #14 of Table 8B);

[0412] CLNK rs6846529, CLU rsll787077, and ABCA7 rsl2151021 (occurring in line #16 of Table 8B);

[0413] HLAdqal rs6605556, EED rs3851179, and D0C2A rsll40239 (occurring in line #17 of Table 8B);

[0414] HLAdqal rs6605556, CLU rsll787077, and KLF16 rsl49080927 (occurring in line #20 of Table 8B);

[0415] CLNK rs6846529, EPHA1 rsll771145, and CLU rsll787077 (occurring in lines #21 and 22 of Table 8B in different combinations of number of risk alleles per SGV);

[0416] COX7C rs62374257, EED rs3851179, and TPCN1 rs6489896 (occurring in line #23 of Table 8B);

[0417] CLU rsll787077, EED rs3851179, and WDR81 rs35048651 (occurring in line #24 of Table 8B).

[0418] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 30, i.e. the combinations listed in Table 8B in lines #1 to and including #22. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being BINI rs6733839, these combinations further including either of the following triplets of SGVs:

[0419] RASGEF1C rsll3706587, HLAdqal rs6605556, and EED rs3851179 (occurring in line #1 of Table 8B);

[0420] HLAdqal rs6605556, EED rs3851179, and MAF rs450674 (occurring in lines #2, 5, and 7 of Table 8B in different combinations of number of risk alleles per SGV);

[0421] HLAdqal rs6605556, MS4A4A rsl582763, and ABCA7 rsl2151021 (occurring in line #3 of Table 8B); HLAdqal rs6605556, MINDY2 rs602602, and ADAMTS1 rs2830489 (occurring in line #4 of Table 8B); HLAdqal rs6605556, UMAD1 rs6943429, and MINDY2 rs602602 (occurring in lines #6, 10, 13 and 18 of Table 8B in different combinations of number of risk alleles per SGV);

[0422] HLAdqal rs6605556, CLU rsll787077, and EED rs3851179 (occurring in line #8 of Table 8B);

[0423] CR1 rs679515, EED rs3851179, and TPCN1 rs6489896 (occurring in line #9 of Table 8B);

[0424] SPI1 rsl0437655, EED rs3851179, and RBCK1 rsl358782 (occurring in lines #11 and 19 of Table 8B in different combinations of number of risk alleles per SGV);

[0425] HLAdqal rs6605556, EED rs3851179, and SNX1 rs3848143 (occurring in lines #12 and 15 of Table 8B in different combinations of number of risk alleles per SGV);

[0426] HS3ST5 rs785129, EPHA1 rsll771145, and EED rs3851179 (occurring in line #14 of Table 8B);

[0427] CLNK rs6846529, CLU rsll787077, and ABCA7 rsl2151021 (occurring in line #16 of Table 8B);

[0428] HLAdqal rs6605556, EED rs3851179, and DOC2A rsll40239 (occurring in line #17 of Table 8B); HLAdqal rs6605556, CLU rsll787077, and KLF16 rsl49080927 (occurring in line #20 of Table 8B);

[0429] CLNK rs6846529, EPHA1 rsll771145, and CLU rsll787077 (occurring in lines #21 and 22 of Table 8B in different combinations of number of risk alleles per SGV).

[0430] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 35, i.e. the combinations listed in Table 8B in lines #1 to and including #20. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being BINI rs6733839, these combinations further including either of the following triplets of SGVs:

[0431] RASGEF1C rsll3706587, HLAdqal rs6605556, and EED rs3851179 (occurring in line #1 of Table 8B);

[0432] HLAdqal rs6605556, EED rs3851179, and MAF rs450674 (occurring in lines #2, 5, and 7 of Table 8B in different combinations of number of risk alleles per SGV);

[0433] HLAdqal rs6605556, MS4A4A rsl582763, and ABCA7 rsl2151021 (occurring in line #3 of Table 8B); HLAdqal rs6605556, MINDY2 rs602602, and ADAMTS1 rs2830489 (occurring in line #4 of Table 8B); HLAdqal rs6605556, UMAD1 rs6943429, and MINDY2 rs602602 (occurring in lines #6, 10, 13 and 18 of Table 8B in different combinations of number of risk alleles per SGV);

[0434] HLAdqal rs6605556, CLU rsll787077, and EED rs3851179 (occurring in line #8 of Table 8B);

[0435] CR1 rs679515, EED rs3851179, and TPCN1 rs6489896 (occurring in line #9 of Table 8B);

[0436] SPI1 rsl0437655, EED rs3851179, and RBCK1 rsl358782 (occurring in lines #11 and 19 of Table 8B in different combinations of number of risk alleles per SGV);

[0437] HLAdqal rs6605556, EED rs3851179, and SNX1 rs3848143 (occurring in lines #12 and 15 of Table 8B in different combinations of number of risk alleles per SGV);

[0438] HS3ST5 rs785129, EPHA1 rsll771145, and EED rs3851179 (occurring in line #14 of Table 8B);

[0439] CLNK rs6846529, CLU rsll787077, and ABCA7 rsl2151021 (occurring in line #16 of Table 8B);

[0440] HLAdqal rs6605556, EED rs3851179, and DOC2A rsll40239 (occurring in line #17 of Table 8B); HLAdqal rs6605556, CLU rsll787077, and KLF16 rsl49080927 (occurring in line #20 of Table 8B).

[0441] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 40, i.e. the combinations listed in Table 8B in lines #1 to and including #19. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being BINI rs6733839, these combinations further including either of the following triplets of SGVs:

[0442] RASGEF1C rsll3706587, HLAdqal rs6605556, and EED rs3851179 (occurring in line #1 of Table 8B);

[0443] HLAdqal rs6605556, EED rs3851179, and MAF rs450674 (occurring in lines #2, 5, and 7 of Table 8B in different combinations of number of risk alleles per SGV); HLAdqal rs6605556, MS4A4A rsl582763, and ABCA7 rsl2151021 (occurring in line #3 of Table 8B); HLAdqal rs6605556, MINDY2 rs602602, and ADAMTS1 rs2830489 (occurring in line #4 of Table 8B); HLAdqal rs6605556, UMAD1 rs6943429, and MINDY2 rs602602 (occurring in lines #6, 10, 13 and 18 of Table 8B in different combinations of number of risk alleles per SGV);

[0444] HLAdqal rs6605556, CLU rsll787077, and EED rs3851179 (occurring in line #8 of Table 8B);

[0445] CR1 rs679515, EED rs3851179, and TPCN1 rs6489896 (occurring in line #9 of Table 8B);

[0446] SPI1 rsl0437655, EED rs3851179, and RBCK1 rsl358782 (occurring in lines #11 and 19 of Table 8B in different combinations of number of risk alleles per SGV);

[0447] HLAdqal rs6605556, EED rs3851179, and SNX1 rs3848143 (occurring in lines #12 and 15 of Table 8B in different combinations of number of risk alleles per SGV);

[0448] HS3ST5 rs785129, EPHA1 rsll771145, and EED rs3851179 (occurring in line #14 of Table 8B); CLNK rs6846529, CLU rsll787077, and ABCA7 rsl2151021 (occurring in line #16 of Table 8B); HLAdqal rs6605556, EED rs3851179, and DOC2A rsll40239 (occurring in line #17 of Table 8B).

[0449] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 50, i.e. the combinations listed in Table 8B in lines #1 to and including #18. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being BINI rs6733839, these combinations further including either of the following triplets of SGVs:

[0450] RASGEF1C rsll3706587, HLAdqal rs6605556, and EED rs3851179 (occurring in line #1 of Table 8B);

[0451] HLAdqal rs6605556, EED rs3851179, and MAF rs450674 (occurring in lines #2, 5, and 7 of Table 8B in different combinations of number of risk alleles per SGV);

[0452] HLAdqal rs6605556, MS4A4A rsl582763, and ABCA7 rsl2151021 (occurring in line #3 of Table 8B); HLAdqal rs6605556, MINDY2 rs602602, and ADAMTS1 rs2830489 (occurring in line #4 of Table 8B); HLAdqal rs6605556, UMAD1 rs6943429, and MINDY2 rs602602 (occurring in lines #6, 10, 13 and 18 of Table 8B in different combinations of number of risk alleles per SGV);

[0453] HLAdqal rs6605556, CLU rsll787077, and EED rs3851179 (occurring in line #8 of Table 8B);

[0454] CR1 rs679515, EED rs3851179, and TPCN1 rs6489896 (occurring in line #9 of Table 8B);

[0455] SPI1 rsl0437655, EED rs3851179, and RBCK1 rsl358782 (occurring in line #11 of Table 8B);

[0456] HLAdqal rs6605556, EED rs3851179, and SNX1 rs3848143 (occurring in lines #12 and 15 of Table 8B in different combinations of number of risk alleles per SGV);

[0457] HS3ST5 rs785129, EPHA1 rsll771145, and EED rs3851179 (occurring in line #14 of Table 8B);

[0458] CLNK rs6846529, CLU rsll787077, and ABCA7 rsl2151021 (occurring in line #16 of Table 8B);

[0459] HLAdqal rs6605556, EED rs3851179, and DOC2A rsll40239 (occurring in line #17 of Table 8B). As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 60, i.e. the combinations listed in Table 8B in lines #1 to and including #17. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being BINI rs6733839, these combinations further include either of the following triplets of SGVs:

[0460] RASGEF1C rsll3706587, HLAdqal rs6605556, and EED rs3851179 (occurring in line #1 of Table 8B);

[0461] HLAdqal rs6605556, EED rs3851179, and MAF rs450674 (occurring in lines #2, 5, and 7 of Table 8B in different combinations of number of risk alleles per SGV);

[0462] HLAdqal rs6605556, MS4A4A rsl582763, and ABCA7 rsl2151021 (occurring in line #3 of Table 8B); HLAdqal rs6605556, MINDY2 rs602602, and ADAMTS1 rs2830489 (occurring in line #4 of Table 8B); HLAdqal rs6605556, UMAD1 rs6943429, and MINDY2 rs602602 (occurring in lines #6, 10 and 13 of Table 8B in different combinations of number of risk alleles per SGV);

[0463] HLAdqal rs6605556, CLU rsll787077, and EED rs3851179 (occurring in line #8 of Table 8B);

[0464] CR1 rs679515, EED rs3851179, and TPCN1 rs6489896 (occurring in line #9 of Table 8B);

[0465] SPI1 rsl0437655, EED rs3851179, and RBCK1 rsl358782 (occurring in line #11 of Table 8B);

[0466] HLAdqal rs6605556, EED rs3851179, and SNX1 rs3848143 (occurring in lines #12 and 15 of Table 8B in different combinations of number of risk alleles per SGV);

[0467] HS3ST5 rs785129, EPHA1 rsll771145, and EED rs3851179 (occurring in line #14 of Table 8B); CLNK rs6846529, CLU rsll787077, and ABCA7 rsl2151021 (occurring in line #16 of Table 8B); HLAdqal rs6605556, EED rs3851179, and DOC2A rsll40239 (occurring in line #17 of Table 8B).

[0468] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 70, i.e. the combinations listed in Table 8B in lines #1 to and including #15. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being BINI rs6733839, these combinations further including either of the following triplets of SGVs:

[0469] RASGEF1C rsll3706587, HLAdqal rs6605556, and EED rs3851179 (occurring in line #1 of Table 8B);

[0470] HLAdqal rs6605556, EED rs3851179, and MAF rs450674 (occurring in lines #2, 5, and 7 of Table 8B in different combinations of number of risk alleles per SGV);

[0471] HLAdqal rs6605556, MS4A4A rsl582763, and ABCA7 rsl2151021 (occurring in line #3 of Table 8B); HLAdqal rs6605556, MINDY2 rs602602, and ADAMTS1 rs2830489 (occurring in line #4 of Table 8B); HLAdqal rs6605556, UMAD1 rs6943429, and MINDY2 rs602602 (occurring in lines #6, 10 and 13 of Table 8B in different combinations of number of risk alleles per SGV);

[0472] HLAdqal rs6605556, CLU rsll787077, and EED rs3851179 (occurring in line #8 of Table 8B);

[0473] CR1 rs679515, EED rs3851179, and TPCN1 rs6489896 (occurring in line #9 of Table 8B); SPI1 rsl0437655, EED rs3851179, and RBCK1 rsl358782 (occurring in line #11 of Table 8B);

[0474] HLAdqal rs6605556, EED rs3851179, and SNX1 rs3848143 (occurring in lines #12 and 15 of Table 8B in different combinations of number of risk alleles per SGV);

[0475] HS3ST5 rs785129, EPHA1 rsll771145, and EED rs3851179 (occurring in line #14 of Table 8B).

[0476] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 80, i.e. the combinations listed in Table 8B in lines #1 to and including #13. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being BINI rs6733839, these combinations further include either of the following triplets of SGVs:

[0477] RASGEF1C rsll3706587, HLAdqal rs6605556, and EED rs3851179 (occurring in line #1 of Table 8B);

[0478] HLAdqal rs6605556, EED rs3851179, and MAF rs450674 (occurring in lines #2, 5, and 7 of Table 8B in different combinations of number of risk alleles per SGV);

[0479] HLAdqal rs6605556, MS4A4A rsl582763, and ABCA7 rsl2151021 (occurring in line #3 of Table 8B); HLAdqal rs6605556, MINDY2 rs602602, and ADAMTS1 rs2830489 (occurring in line #4 of Table 8B); HLAdqal rs6605556, UMAD1 rs6943429, and MINDY2 rs602602 (occurring in lines #6, 10 and 13 of Table 8B in different combinations of number of risk alleles per SGV);

[0480] HLAdqal rs6605556, CLU rsll787077, and EED rs3851179 (occurring in line #8 of Table 8B);

[0481] CR1 rs679515, EED rs3851179, and TPCN1 rs6489896 (occurring in line #9 of Table 8B);

[0482] SPI1 rsl0437655, EED rs3851179, and RBCK1 rsl358782 (occurring in line #11 of Table 8B);

[0483] HLAdqal rs6605556, EED rs3851179, and SNX1 rs3848143 (occurring in line #12 of Table 8B).

[0484] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 90, i.e. the combinations listed in Table 8B in lines #1 to and including #12. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being BINI rs6733839, these combinations further including either of the following triplets of SGVs:

[0485] RASGEF1C rsll3706587, HLAdqal rs6605556, and EED rs3851179 (occurring in line #1 of Table 8B);

[0486] HLAdqal rs6605556, EED rs3851179, and MAF rs450674 (occurring in lines #2, 5, and 7 of Table 8B in different combinations of number of risk alleles per SGV);

[0487] HLAdqal rs6605556, MS4A4A rsl582763, and ABCA7 rsl2151021 (occurring in line #3 of Table 8B); HLAdqal rs6605556, MINDY2 rs602602, and ADAMTS1 rs2830489 (occurring in line #4 of Table 8B); HLAdqal rs6605556, UMAD1 rs6943429, and MINDY2 rs602602 (occurring in lines #6 and 10 of Table 8B in different combinations of number of risk alleles per SGV);

[0488] HLAdqal rs6605556, CLU rsll787077, and EED rs3851179 (occurring in line #8 of Table 8B); CR1 rs679515, EED rs3851179, and TPCN1 rs6489896 (occurring in line #9 of Table 8B);

[0489] SPI1 rsl0437655, EED rs3851179, and RBCK1 rsl358782 (occurring in line #11 of Table 8B);

[0490] HLAdqal rs6605556, EED rs3851179, and SNX1 rs3848143 (occurring in line #12 of Table 8B).

[0491] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 100, i.e. the combinations listed in Table 8B in lines #1 to and including #11. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being BINI rs6733839, these combinations further including either of the following triplets of SGVs:

[0492] RASGEF1C rsll3706587, HLAdqal rs6605556, and EED rs3851179 (occurring in line #1 of Table 8B);

[0493] HLAdqal rs6605556, EED rs3851179, and MAF rs450674 (occurring in lines #2, 5, and 7 of Table 8B in different combinations of number of risk alleles per SGV);

[0494] HLAdqal rs6605556, MS4A4A rsl582763, and ABCA7 rsl2151021 (occurring in line #3 of Table 8B); HLAdqal rs6605556, MINDY2 rs602602, and ADAMTS1 rs2830489 (occurring in line #4 of Table 8B); HLAdqal rs6605556, UMAD1 rs6943429, and MINDY2 rs602602 (occurring in lines #6 and 10 of Table 8B in different combinations of number of risk alleles per SGV);

[0495] HLAdqal rs6605556, CLU rsll787077, and EED rs3851179 (occurring in line #8 of Table 8B);

[0496] CR1 rs679515, EED rs3851179, and TPCN1 rs6489896 (occurring in line #9 of Table 8B);

[0497] SPI1 rsl0437655, EED rs3851179, and RBCK1 rsl358782 (occurring in line #11 of Table 8B).

[0498] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 120, i.e. the combinations listed in Table 8B in lines #1 to and including #9. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being BINI rs6733839, these combinations further including either of the following triplets of SGVs:

[0499] RASGEF1C rsll3706587, HLAdqal rs6605556, and EED rs3851179 (occurring in line #1 of Table 8B);

[0500] HLAdqal rs6605556, EED rs3851179, and MAF rs450674 (occurring in lines #2, 5, and 7 of Table 8B in different combinations of number of risk alleles per SGV);

[0501] HLAdqal rs6605556, MS4A4A rsl582763, and ABCA7 rsl2151021 (occurring in line #3 of Table 8B); HLAdqal rs6605556, MINDY2 rs602602, and ADAMTS1 rs2830489 (occurring in line #4 of Table 8B); HLAdqal rs6605556, UMAD1 rs6943429, and MINDY2 rs602602 (occurring in line #6 of Table 8B); HLAdqal rs6605556, CLU rsll787077, and EED rs3851179 (occurring in line #8 of Table 8B);

[0502] CR1 rs679515, EED rs3851179, and TPCN1 rs6489896 (occurring in line #9 of Table 8B). As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 140, i.e. the combinations listed in Table 8B in lines #1 to and including #7. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being BINI rs6733839, these combinations further including either of the following triplets of SGVs:

[0503] RASGEF1C rsll3706587, HLAdqal rs6605556, and EED rs3851179 (occurring in line #1 of Table 8B);

[0504] HLAdqal rs6605556, EED rs3851179, and MAF rs450674 (occurring in lines #2, 5, and 7 of Table 8B in different combinations of number of risk alleles per SGV);

[0505] HLAdqal rs6605556, MS4A4A rsl582763, and ABCA7 rsl2151021 (occurring in line #3 of Table 8B); HLAdqal rs6605556, MINDY2 rs602602, and ADAMTS1 rs2830489 (occurring in line #4 of Table 8B); HLAdqal rs6605556, UMAD1 rs6943429, and MINDY2 rs602602 (occurring in line #6 of Table 8B).

[0506] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination with an odds ratio equal to or higher than 160, i.e. the combinations listed in Table 8B in lines #1 to and including #5. More in particular, these are combinations of 4 SGVs with one of the 4 SGVs being BINI rs6733839, these combinations further including either of the following triplets of SGVs:

[0507] RASGEF1C rsll3706587, HLAdqal rs6605556, and EED rs3851179 (occurring in line #1 of Table 8B);

[0508] HLAdqal rs6605556, EED rs3851179, and MAF rs450674 (occurring in lines #2, and 5 of Table 8B in different combinations of number of risk alleles per SGV);

[0509] HLAdqal rs6605556, MS4A4A rsl582763, and ABCA7 rsl2151021 (occurring in line #3 of Table 8B); HLAdqal rs6605556, MINDY2 rs602602, and ADAMTS1 rs2830489 (occurring in line #4 of Table 8B).

[0510] As such, in particular, a combination of 4 SGVs is a combination of the 4 SGVs BINI rs6733839, HLAdqal rs6605556, MS4A4A rsl582763, and ABCA7 rsl2151021 (occurring in line #3 of Table 8B).

[0511] As such, in particular, a combination of 4 SGVs is a combination of the 4 SGVs BINI rs6733839, HLAdqal rs6605556, EED rs3851179, and MAF rs450674 (occurring in lines #2, and 5 of Table 8B).

[0512] As such, in particular, a combination of 4 SGVs is a combination of the 4 SGVs BINI rs6733839, RASGEF1C rsll3706587, HLAdqal rs6605556, and EED rs3851179 (occurring in line #1 of Table 8B).

[0513] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs including the SGV APOE rs429358 is a combination comprising the SGV APOE rs429358 and a further SGV with a frequency of occurrence (Table 7C) of at least 2 out of the 31 withheld combinations. More in particular, such combinations of 4 SGVs are comprising the SGV APOE rs429358 and at least one of the SGVs chosen from: SHARPIN rs34173062, COX7C rs62374257, BINI rs6733839, PTK2B rs73223431, RASGEF1C rsll3706587, SNX1 rs3848143, IGHcluster rs7157106, MY015A rs2242595, KLF16 rsl49080927, RHOH rs2245466, SCIMP rs7225151, APOE rs7412, EPDR1 rs6966331, GRN rs5848, LILRB2 rs587709, MME rs61762319, PLEKHA1 rs7908662, SLC24A4 rs7401792, and TSPOAP1 rs2526377. In one embodiment thereto, the SGV APOE rs429358 and the chosen SGV are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV APOE rs429358 and different from the chosen SGV.

[0514] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs including the SGV APOE rs429358 is a combination comprising the SGV APOE rs429358 and a further SGV with a frequency of occurrence (Table 7C) of at least 3 out of the 31 withheld combinations. More in particular, such combinations of 4 SGVs are comprising the SGV APOE rs429358 and at least one of the SGVs chosen from: SHARPIN rs34173062, COX7C rs62374257, BINI rs6733839, PTK2B rs73223431, RASGEF1C rsll3706587, SNX1 rs3848143, IGHcluster rs7157106, MY015A rs2242595, KLF16 rsl49080927, RHOH rs2245466, and SCIMP rs7225151. In one embodiment thereto, the SGV APOE rs429358 and the chosen SGV are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV APOE rs429358 and different from the chosen SGV.

[0515] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs including the SGV APOE rs429358 is a combination comprising the SGV APOE rs429358 and a further SGV with a frequency of occurrence (Table 7C) of at least 4 out of the 31 withheld combinations. More in particular, such combinations of 4 SGVs are comprising the SGV APOE rs429358 and at least one of the SGVs chosen from: SHARPIN rs34173062, COX7C rs62374257, BINI rs6733839, PTK2B rs73223431, RASGEF1C rsll3706587, SNX1 rs3848143, IGHcluster rs7157106, and MY015A rs2242595. In one embodiment thereto, the SGV APOE rs429358 and the chosen SGV are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV APOE rs429358 and different from the chosen SGV.

[0516] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs including the SGV APOE rs429358 is a combination comprising the SGV APOE rs429358 and a further SGV with a frequency of occurrence (Table 7C) of at least 5out of the 31 withheld combinations. More in particular, such combinations of 4 SGVs are comprising the SGV APOE rs429358 and at least one of the SGVs chosen from SHARPIN rs34173062, COX7C rs62374257, BINI rs6733839, PTK2B rs73223431, RASGEF1C rsll3706587, and SNX1 rs3848143. In one embodiment thereto, the SGV APOE rs429358 and the chosen SGV are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV APOE rs429358 and different from the chosen SGV.

[0517] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs including the SGV APOE rs429358 is a combination comprising the SGV COX7C rs62374257. More in particular, such combinations of 4 SGVs are comprising the SGV APOE rs429358 and the SGV COX7C rs62374257. In one embodiment thereto, the SGV APOE rs429358 and the SGV COX7C rs62374257 are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV APOE rs429358 and different from the SGV COX7C rs62374257.

[0518] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs including the SGV APOE rs429358 is a combination comprising the SGV SHARPIN rs34173062. More in particular, such combinations of 4 SGVs are comprising the SGV APOE rs429358 and the SGV SHARPIN rs34173062. In one embodiment thereto, the SGV APOE rs429358 and the SGV SHARPIN rs34173062 are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV APOE rs429358 and different from the SGV SHARPIN rs34173062.

[0519] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs including the SGV APOE rs429358 is a combination comprising the SGV APOE rs429358 and a further SGV chosen from ANK3 rs7068231, CLNK rs6846529, DOC2A rsll40239, HS3ST5 rs785129, MS4A4A rsl582763, TNIP1 rs871269, and TPCN1 rs6489896. More in particular, such combinations of 4 SGVs are comprising the SGV APOE rs429358 and at least one of the SGVs chosen from ANK3 rs7068231, CLNK rs6846529, DOC2A rsll40239, HS3ST5 rs785129, MS4A4A rsl582763, TNIP1 rs871269, and TPCN1 rs6489896. In one embodiment thereto, the SGV APOE rs429358 and the chosen SGV are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV APOE rs429358 and different from the chosen SGV.

[0520] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination comprising the SGV BINI rs6733839 and one further SGV with a frequency of occurrence (Table 8C) of at least 4 out of the 120 withheld combinations. More in particular, such combinations of 4 SGVs are comprising the SGV BINI rs6733839 and at least one of the SGVs chosen from EED rs3851179, CLU rsll787077, EPHA1 rsll771145, ABCA7 rsl2151021, MINDY2 rs602602, CLNK rs6846529, HLAdqal rs6605556, SPI1 rsl0437655, IDUA rs3822030, HS3ST5 rs785129, CR1 rs679515, UMAD1 rs6943429, COX7C rs62374257, SNX1 rs3848143, SLC2A4RG rs6742, ABCA1 rsl800978, MAF rs450674, PLEKHA1 rs7908662, PTK2B rs73223431, ANK3 rs7068231, and RHOH rs2245466. In one embodiment thereto, the SGV BINI rs6733839 and the chosen SGV are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV BINI rs6733839 and different from the chosen SGV.

[0521] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination comprising the SGV BINI rs6733839 and one further SGV with a frequency of occurrence (Table 8C) of at least 5 out of the 120 withheld combinations. More in particular, such combinations of 4 SGVs are comprising the SGV BINI rs6733839 and at least one of the SGVs chosen from EED rs3851179, CLU rsll787077, EPHA1 rsll771145, ABCA7 rsl2151021, MINDY2 rs602602, CLNK rs6846529, HLAdqal rs6605556, SPI1 rsl0437655, IDUA rs3822030, HS3ST5 rs785129, CR1 rs679515, UMAD1 rs6943429, COX7C rs62374257, SNX1 rs3848143, SLC2A4RG rs6742, ABCA1 rsl800978, MAF rs450674, PLEKHA1 rs7908662, and PTK2B rs73223431. In one embodiment thereto, the SGV BINI rs6733839 and the chosen SGV are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV BINI rs6733839 and different from the chosen SGV.

[0522] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination comprising the SGV BINI rs6733839 and one further SGV with a frequency of occurrence (Table 8C) of at least 6 out of the 120 withheld combinations. More in particular, such combinations of 4 SGVs are comprising the SGV BINI rs6733839 and at least one of the SGVs chosen from EED rs3851179, CLU rsll787077, EPHA1 rsll771145, ABCA7 rsl2151021, MINDY2 rs602602, CLNK rs6846529, HLAdqal rs6605556, SPI1 rsl0437655, IDUA rs3822030, HS3ST5 rs785129, CR1 rs679515, UMAD1 rs6943429, COX7C rs62374257, SNX1 rs3848143, and SLC2A4RG rs6742. In one embodiment thereto, the SGV BINI rs6733839 and the chosen SGV are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV BINI rs6733839 and different from the chosen SGV.

[0523] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination comprising the SGV BINI rs6733839 and one further SGV with a frequency of occurrence (Table 8C) of at least 7 out of the 120 withheld combinations. More in particular, such combinations of 4 SGVs are comprising the SGV BINI rs6733839 and at least one of the SGVs chosen from EED rs3851179, CLU rsll787077, EPHA1 rsll771145, ABCA7 rsl2151021, MINDY2 rs602602, CLNK rs6846529, HLAdqal rs6605556, SPI1 rsl0437655, IDUA rs3822030, HS3ST5 rs785129, CR1 rs679515, UMAD1 rs6943429, COX7C rs62374257, and SNX1 rs3848143. In one embodiment thereto, the SGV BINI rs6733839 and the chosen SGV are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV BINI rs6733839 and different from the chosen SGV.

[0524] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination comprising the SGV BINI rs6733839 and one further SGV with a frequency of occurrence (Table 8C) of at least 8 out of the 120 withheld combinations. More in particular, such combinations of 4 SGVs are comprising the SGV BINI rs6733839 and at least one of the SGVs chosen from EED rs3851179, CLU rsll787077, EPHA1 rsll771145, ABCA7 rsl2151021, MINDY2 rs602602, CLNK rs6846529, HLAdqal rs6605556, SPI1 rsl0437655, IDUA rs3822030, HS3ST5 rs785129, CR1 rs679515, and UMAD1 rs6943429. In one embodiment thereto, the SGV BINI rs6733839 and the chosen SGV are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV BINI rs6733839 and different from the chosen SGV.

[0525] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination comprising the SGV BINI rs6733839 and one further SGV with a frequency of occurrence (Table 8C) of at least 9 out of the 120 withheld combinations. More in particular, such combinations of 4 SGVs are comprising the SGV BINI rs6733839 and at least one of the SGVs chosen from EED rs3851179, CLU rsll787077, EPHA1 rsll771145, ABCA7 rsl2151021, MINDY2 rs602602, CLNK rs6846529, HLAdqal rs6605556, SPI1 rsl0437655, IDUA rs3822030, and HS3ST5 rs785129. In one embodiment thereto, the SGV BINI rs6733839 and the chosen SGV are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV BINI rs6733839 and different from the chosen SGV.

[0526] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination comprising the SGV BINI rs6733839 and one further SGV with a frequency of occurrence (Table 8C) of at least 11 out of the 120 withheld combinations. More in particular, such combinations of 4 SGVs are comprising the SGV BINI rs6733839 and at least one of the SGVs chosen from EED rs3851179, CLU rsll787077, EPHA1 rsll771145, ABCA7 rsl2151021, MINDY2 rs602602, CLNK rs6846529, HLAdqal rs6605556, SPI1 rsl0437655, and IDUA rs3822030. In one embodiment thereto, the SGV BINI rs6733839 and the chosen SGV are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV BINI rs6733839 and different from the chosen SGV.

[0527] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination comprising the SGV BINI rs6733839 and one further SGV with a frequency of occurrence (Table 8C) of at least 20 out of the 120 withheld combinations. More in particular, such combinations of 4 SGVs are comprising the SGV BINI rs6733839 and at least one of the SGVs chosen from EED rs3851179, CLU rsll787077, EPHA1 rsll771145, ABCA7 rsl2151021, MINDY2 rs602602, CLNK rs6846529, HLAdqal rs6605556, and SPI1 rsl0437655. In one embodiment thereto, the SGV BINI rs6733839 and the chosen SGV are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV BINI rs6733839 and different from the chosen SGV.

[0528] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination comprising the SGV BINI rs6733839 and the further SGV HLAdqal rs6605556. More in particular, such combinations of 4 SGVs are comprising the SGV BINI rs6733839 and the SGV HLAdqal rs6605556. In one embodiment thereto, the SGV BINI rs6733839 and the SGV HLAdqal rs6605556 are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV BINI rs6733839 and different from SGV HLAdqal rs6605556.

[0529] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination comprising the SGV BINI rs6733839 and the further SGV CLNK rs6846529. More in particular, such combinations of 4 SGVs are comprising the SGV BINI rs6733839 and the SGV CLNK rs6846529. In one embodiment thereto, the SGV BINI rs6733839 and the SGV CLNK rs6846529 are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV BINI rs6733839 and different from SGV CLNK rs6846529.

[0530] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination comprising the SGV BINI rs6733839 and the further SGV MINDY2 rs602602. More in particular, such combinations of 4 SGVs are comprising the SGV BINI rs6733839 and the SGV MINDY2 rs602602. In one embodiment thereto, the SGV BINI rs6733839 and the SGV MINDY2 rs602602 are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV BINI rs6733839 and different from SGV MINDY2 rs602602.

[0531] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination comprising the SGV BINI rs6733839 and the further SGV ABCA7 rsl2151021. More in particular, such combinations of 4 SGVs are comprising the SGV BINI rs6733839 and the SGV ABCA7 rsl2151021. In one embodiment thereto, the SGV BINI rs6733839 and the SGV ABCA7 rsl2151021 are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV BINI rs6733839 and different from SGV ABCA7 rsl2151021.

[0532] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination comprising the SGV BINI rs6733839 and the further SGV EPHA1 rsll771145. More in particular, such combinations of 4 SGVs are comprising the SGV BINI rs6733839 and the SGV EPHA1 rsll771145. In one embodiment thereto, the SGV BINI rs6733839 and the SGV EPHA1 rsll771145 are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV BINI rs6733839 and different from SGV EPHA1 rsll771145.

[0533] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination comprising the SGV BINI rs6733839 and the further SGV CLU rsll787077. More in particular, such combinations of 4 SGVs are comprising the SGV BINI rs6733839 and the SGV CLU rsll787077. In one embodiment thereto, the SGV BINI rs6733839 and the SGV CLU rsll787077 are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV BINI rs6733839 and different from SGV CLU rsll787077.

[0534] As such, in particular, a combination of 4 SGVs from the withheld combinations of 4 SGVs excluding the SGV APOE rs429358 is a combination comprising the SGV BINI rs6733839 and the further SGV EED rs3851179. More in particular, such combinations of 4 SGVs are comprising the SGV BINI rs6733839 and the SGV EED rs3851179. In one embodiment thereto, the SGV BINI rs6733839 and the SGV EED rs3851179 are combined with a third SGV and a fourth SGV of Table 1 wherein the third SGV and fourth SGV each are different from the SGV BINI rs6733839 and different from SGV EED rs3851179.

[0535] Any of the above methods of separating / distinguishing AD patients or subjects at risk of developing AD from healthy / control subjects, or of diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or as potential AD patient, or as subject at risk of developing AD, can be applied in a number of applications, such as: methods of screening for AD drugs / medicaments, e.g. in vitro screening for AD drug / medicaments using cultured diseased cells from AD patients, such as induced pluripotent stem cells produced from AD patients (e.g. Kondo et al. 2017, Cell Reports 21: 2304-2312), wherein the AD patients can be selected using the methods as described herein; selection of subjects for inclusion in a clinical trial of a candidate AD drug / medicament or candidate drug / medicament for treating AD, wherein the subjects can be selected using the methods as described herein; selection of subjects for treatment with a drug / medicament, in particular an AD drug / medicament, wherein the AD patients can be selected using the methods as described herein; in particular with progression in this field it is not unlikely that AD patients will the future (need to) be stratified in and that available medicaments for treating AD patients will be more effective in a subset of AD patients.

[0536] This disclosure therefore further relates to methods of screening, in particular in vitro screening, for AD drugs / medicaments, the methods comprising: culturing cells obtained from an AD patient, wherein the AD patient is selected by means of / using / applying / based on any methods of separating / distinguishing AD patients or subjects at risk of developing AD from healthy / control subjects, or on any methods of diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or as potential AD patient, or as subject at risk of developing AD, as described herein, and wherein the cultured cells are displaying an AD-related phenotype; contacting the cultured cells with a candidate AD drug / medicament; selecting as AD drug / medicament a candidate AD drug / medicament that is treating the AD- related phenotype.

[0537] This disclosure therefore further relates to methods of selecting a subject or subjects for inclusion in a clinical trial of a candidate AD drug / medicament or candidate drug / medicament for treating AD, wherein the subjects are selected by means of / using / applying / based on any methods of separating / distinguishing AD patients or subjects at risk of developing AD from healthy / control subjects, or on any methods of diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or as potential AD patient, or as subject at risk of developing AD, as described herein.

[0538] This disclosure therefore further relates to methods of selecting a subject or subjects for treatment with a drug / medicament, in particular an AD drug / medicament, wherein the subject is selected by means of / using / applying / based on any methods of separating / distinguishing AD patients or subjects at risk of developing AD from healthy / control subjects, or on any methods of diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or as potential AD patient, or as subject at risk of developing AD, as described herein.

[0539] The current disclosure further relates to a SGV combination or a combination / set of more than 1 / at least 2 SGV combinations chosen from the SGV combinations as listed in Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table IDA, or Table 11A in, or for use in: - a method of separating / distinguishing AD patients or subjects at risk of developing AD from healthy / control subjects, or a method of diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or as potential AD patient, or as subject at risk of developing AD, as described hereinabove;

[0540] - a method of screening for AD drugs / medicaments as describe hereinabove;

[0541] - a method of selecting a subject for inclusion in a clinical trial of a candidate AD drug / medicament as described hereinabove; or

[0542] - a method of selecting a subject for treatment with a drug / medicament as described hereinabove.

[0543] The current disclosure further relates to use of a SGV combination or a combination / set of more than 1 SGV combination chosen from the SGV combinations as listed in Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table IDA, or Table 11A in:

[0544] - a method of separating / distinguishing AD patients or subjects at risk of developing AD from healthy / control subjects, or a method of diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or as potential AD patient, or as subject at risk of developing AD, as described hereinabove;

[0545] - a method of screening for AD drugs / medicaments as describe hereinabove;

[0546] - a method of selecting a subject for inclusion in a clinical trial of a candidate AD drug / medicament as described hereinabove; or

[0547] - a method of selecting a subject for treatment with a drug / medicament as described hereinabove.

[0548] Many regions of the genome are inherited together (independent of whether physically linked on e.g. the same chromosome or not). The likelihood for two genetic variants, such as an individual SGV as described herein (pathogenic SGV, SGV linked to AD, also referred to as AD-SGV) and another SGV to be inherited together is measured by Linkage Disequilibrium (LD) analysis, which basically provides a score between 0 (independent loci, inherited separately) and 1 (dependent loci, inherited together). LD analysis was performed for each of the individual AD-SGVs, yielding a list of SGVs in LD (also referred to as "LD-SGVs") with the AD-SGVs. This LD analysis was run on a 1000 Genomes panel for European nonFinnish populations as a reference panel; this implies that performing the same analysis on a different reference panel of human genomes may render a different list of LD-SGVs. In addition, LD-SGVs in the list of Table 12 are further restricted to those (i) with an LD score (r2) equal to or higher than 0.8 and (ii) within a distance of 1000 kilobases from an AD-SGV; this implies that changing either of these parameters will influence the outcome of the analysis, i.e., will change the list of LD-SGVs. Finally, the algorithm applied to determine whether two SGV are in LD may influence the outcome of the LD analysis. The LD-SGVs as listed in Table 12 therefore are exemplary ones but provide solid support for this concept. Of note, the risk SGVs rsl41749679, rs60755019 and rsl800978 of Table 1 are not represented in Table 12 as the risk SGVs rsl41749679 and rs60755019 are not present in the analyzed reference panel whereas the analysis for risk SGV rsl800978 is ongoing. As the LD-SGVs (any, thus not limited to those of Table 12) are in LD with the AD-SGVs, any particular AD-SGV in a combination of 2, 3 or 4 SGVs as described above can be exchanged for an LD-SGVs associated with that particular AD-SGV, this extends to all methods, computer programs, computing or data processing systems or devices etc. as described above.

[0549] The current disclosure therefore further relates to:

[0550] - methods of separating / distinguishing AD patients or subjects at risk of developing AD from healthy / control subjects, or a method of diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or as potential AD patient, or as subject at risk of developing AD, as described hereinabove;

[0551] - methods of screening for AD drugs / medicaments as describe hereinabove;

[0552] - methods of selecting a subject for inclusion in a clinical trial of a candidate AD drug / medicament as described hereinabove; or

[0553] - methods of selecting a subject for treatment with a drug / medicament as described hereinabove; wherein an SGV of a combination or of a combination / set of more than 1 / at least 2 SGV combinations chosen from SGV combinations as listed in Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table IDA, or Table 11A (or particular SGV combinations selected therefrom as outlined hereinabove) is substituted by an SGV that is in linkage disequilibrium (LD) with the SGV of the combination. In particular herein, the SGVs that are in LD with the SGVs of the combinations are listed in Table 12 - more in particular, these SGVs that are in LD with the SGVs of the combinations are determined relative to the phase 3 release of the 1000 Genomes panel for European non-Finnish populations (ftp.1000genomes.ebi.ac.uk / voll / ftp / release / 20130502 / ), wherein the SGVs are in LD within a distance of 1000 kilobases (1 megabase) of the SGVs or the combinations, wherein the LD was calculated in plink 1.9 (cog-genomics.org / plink / 1.9 / ), and wherein the SGVs that are in LD with the SGVs of the combinations have an LD score of r2>=0.8.

[0554] Determining the presence of an SGV

[0555] Polynucleotide sequence determination is a common step in the methods as applied in the current disclosure. When a determined polynucleotide sequence is including or spanning an SGV as described herein (individual SGVs see Table 1), than it can be determined which of the possible alternative nucleotides is present in the DNA obtained from a subject, more in particular in each of the alleles in the DNA obtained from a subject. In general, sequencing methods may include, but are not limited to: high-throughput sequencing, pyrosequencing, sequencing-by-synthesis, single-molecule sequencing, nanopore sequencing, semiconductor sequencing, sequencing-by-ligation, sequencing-by-hybridization, RNA-Seq (Illumina), Digital Gene Expression (Helicos), Next generation sequencing, Single Molecule Sequencing by Synthesis (SMSS) (Helicos), massively- parallel sequencing, Clonal Single Molecule Array (Solexa), shotgun sequencing, Maxam- Gilbert or Sanger sequencing, primer walking, sequencing using PacBio, SOLiD, Ion Torrent, or Nanopore platforms, short read sequencing, long read sequencing, and any other sequencing methods known in the art. The sequencing method can be massively parallel sequencing, that is, simultaneously (or in rapid succession) sequencing any of at least 100, 1000, 10,000, 100,000, 1 million, 10 million, 100 million, 1 billion, or 10 billion polynucleotide molecules.

[0556] Certain DNA sequencing methods may rely on the capture of polynucleotides of interest such as to enrich for these sequences of interest. Polynucleotide or sequence capture typically involves the use of oligonucleotide probes that hybridize to the polynucleotide or sequence of interest. A probe set strategy can involve tiling the probes across a region of interest (complete or partial tiling of the target sequence with probes). Such probes can be, e.g., 10 to 400 or about 400 bases long, 10 to 300 or about 300 bases long, 10 to 200 or about 200 bases long, 10 to 100 or about 100 bases long, 10 to 80 or about 80 bases long, 10 to 60 or about 60 bases long, Such probes may comprise at least one or a set of oligonucleotides of 10 to 60 bases or nucleotides long and / or comprise at least one or a set of oligonucleotides of 15 to 120 bases or nucleotides long. Any set of such oligonucleotide probes can have a depth of about O.lx, 0.2x, 0.3x, 0.4 x, 0.5x, O.lx to 0.5x, lx 2x, 3x, 4x, 5x, 6x, 8x, 9x, lOx, 15x, 20x, 50x or more. Enriched nucleic acid molecules can be representative of a nucleic acid features of interest such as, but not necessarily limited to short genomic variants (SGVs) such as insertions or deletions (e.g., indels) and SNPs.

[0557] Sequencing depth refers to the number of times a locus is covered by a sequence read aligned to the locus. A locus can be as small as a nucleotide, as large as a chromosome arm, or as large as the entire genome. Sequencing depth can be expressed as e.g. lOx, 50x, lOOx, where "x" refers to the number of times a locus is covered by a sequence read. Sequencing depth can also be applied to multiple loci, or to the whole genome, in which case "x" can refer to the mean number of times the loci, or whole genome, is sequenced. Ultra-deep sequencing refers to a sequencing depth of at least lOOx.

[0558] Shallow whole genome sequencing, low coverage whole genome sequencing, or ultra-low pass whole genome sequencing in general refers to short-read sequencing of genomes at low coverage, typically less than 3x coverage, less than 2x coverage, less than lx coverage, such as O.lx to lx coverage, such as O.lx to 0.8x coverage, such as O.lx to 0.6x coverage, such as O.lx to 0.5x coverage, such as O.lx to 0.4x coverage, such as O.lx to 0.3x coverage, such as 0.9x coverage, 0.8x coverage, 0.7x coverage, 0.6x coverage, 0.5x coverage, 0.4x coverage, 0.3x coverage, 0.2x coverage or O.lx coverage, such as O.lx coverage or less. Sequencing coverage can also be expressed as average sequencing coverage. Low coverage in the context of sequencing thus can also refer to typically on average less than 3x coverage, on average less than 2x coverage, on average less than lx coverage, such as on average O.lx to lx coverage, such as on average O.lx to 0.8x coverage, such as on average O.lx to 0.6x coverage, such as on average O.lx to 0.5x coverage, such as on average O.lx to 0.4x coverage, such as on average O.lx to 0.3x coverage, such as on average 0.9x coverage, on average 0.8x coverage, on average 0.7x coverage, on average 0.6x coverage, on average 0.5x coverage, on average 0.4x coverage, on average 0.3x coverage, on average 0.2x coverage or on average O.lx coverage, such as on average O.lx coverage or less.

[0559] By performing shallow whole genome sequencing, low coverage whole genome sequencing, or ultra-low pass whole genome sequencing, each sample is subjected to a small amount of sequencing, allowing application of whole genome sequencing to many samples at low cost per sample.

[0560] A sequence read is a string of nucleotides sequenced from a part or all of a nucleic acid molecule. A sequence read may be a short string of nucleotides (e.g. 20 to 150 nucleotides, around 50 nucleotides) sequenced from a nucleic acid (fragment). Sequence reads may be obtained at one end of a nucleic acid (fragment) or from both ends of a nucleic acid (fragment). Sequence reads may be obtained by e.g. applying a sequencing technique to the nucleic acid (fragment), by hybridization arrays or capture probes, by amplification techniques (e.g. PCR, linear amplification, isothermal amplification) such as amplification techniques using a single primer.

[0561] Thus, obtaining information from the nucleic acid molecules present in a biological sample may include a step of preparing a sequencing library using the nucleic acid molecules isolated from the biological sample. The preparation of such sequencing library may include a step of DNA amplification, or may, alternatively, not include a step of DNA amplification. Obtaining information from the nucleic acid molecules present in a biological sample may include obtaining DNA sequence reads. Obtaining information from the nucleic acid molecules (e.g. cfDNA molecules) present in a biological sample may include the step of aligning the plurality of (DNA) sequence reads to a reference genome to determine the genomic positions of each (individual) sequence read of the plurality of sequence reads. In view of the size of the reference genome and the number of sequence reads in a plurality of sequence reads, the sequence reads are optionally received at a computer system.

[0562] For discriminating closely related sequences, such as SGVs, a number of techniques are available that not necessarily imply full "reading" (i.e. determination of the full sequence) the DNA, or stretches of the DNA of the subject. Other methods allowing the detection of SGVs can comprise a step such as a nucleic acid amplification reaction, a nucleic acid sequencing reaction, a melting reaction, a hybridization reaction to a nucleic acid, or a reverse hybridization reaction to a nucleic acid, or a combination of such steps. Allele-specific PCR, using either primers (allele-specific amplification or allele-specific primer extension; e.g. WO93 / 22456; US5, 137,806; US5,595,890; US5,639,611; US4,851,331) and / or probes (allele-specific hybridization; e.g. Stoneking et al. 1991, Am J Hum Genet 48:70-382; Saiki et al. 1986, Nature 324:163-166; EP235726; WO89 / 11548) is one such technique. Hyperchromicity, intercalating dyes, colorimetric or fluorescent dye detection and fluorescence polarization melting curve analysis are other techniques. Capture probes (see also hereinabove) can be useful in real-time PCR detection of SGVs, and in the detecting SGVs in PCR amplicons using an ELISA-type assays or using microarray analysis. The melting temperatures of two closely related sequences are very similar. In this context, locked nucleic acid (LNA) probes can be helpful in increasing the difference in melting temperatures of a duplex of a LNA probe with its target sequence compared to a duplex of the LNA probe with a variant sequence closely related to its target sequence. LNA primers can be used in allele-specific PCR applications for detecting SGVs. Another method for characterizing SGVs does not require use of a complete PCR, but typically uses only the extension of a primer by a single nucleotide, e.g. a fluorescently-labeled dideoxyribonucleic acid molecule (ddNTP) that is complementary to the nucleotide to be investigated. The nucleotide at the polymorphic site can be identified via detection of a primer that has been extended by one base and is fluorescently labeled (e.g., Kobayashi et al. 1995, Mol Cell Probes 9:175-182).

[0563] Alleles of target sequences can alternatively be differentiated using single-strand conformation polymorphism analysis, which identifies base differences by alteration in electrophoretic migration of single stranded PCR products (e.g. Orita et al. 1989, Proc Natl Acad Sci 86:2766-2770). Amplified PCR products can be generated as described above, and heated or otherwise denatured, to form single stranded amplification products. Single-stranded nucleic acids may refold or form secondary structures which are partially dependent on the base sequence. The different electrophoretic mobilities of singlestranded amplification products can be related to base-sequence difference between alleles of target genes.

[0564] Often one or more artificial, man-made, non-naturally occurring oligonucleotide is used in such methods. In particular, such oligonucleotides can comprise besides ribonucleic acid monomers or deoxyribonucleic acid monomers: one or more modified nucleotide bases, one or more modified nucleotide sugars, one or more labelled nucleotides, one or more peptide nucleic acid monomers, one or more locked nucleic acid monomers, the backbone of such oligonucleotide can be modified, and / or non-glycosidic bonds may link two adjacent nucleotides. Such oligonucleotides may further comprise a modification for attachment to a solid support, e.g., an amine-, thiol-, 3-'propanolamine or acrydite-modification of the oligonucleotide, or may comprise the addition of a homopolymeric tail (for instance an oligo(dT)-ta II added enzymatically via a terminal transferase enzyme or added synthetically) to the oligonucleotide. If said homopolymeric tail is positioned at the 3'-terminus of the oligonucleotide or if any other 3'-terminal modification preventing enzymatic extension is incorporated in the oligonucleotide, the priming capacity of the oligonucleotide can be decreased or abolished. Such oligonucleotides may also comprise a hairpin structure at either end. Terminal extension of such oligonucleotide may be useful for, e.g., specifically hybridizing with another nucleic acid molecule (e.g. when functioning as capture probe), and / or for facilitating attachment of said oligonucleotide to a solid support, and / or for modification of said tailed oligonucleotide by an enzyme, ribozyme or DNAzyme. Such oligonucleotides may be modified in order to detect (the levels of) a target nucleotide sequence and / or to facilitate in any way such detection. Such modifications include labelling with a single label, with two different labels (for instance two fluorophores or one fluorophore and one quencher), the attachment of a different 'universal' tail to two probes or primers hybridizing adjacent or in close proximity to each other with the target nucleotide sequence, the incorporation of a target-specific sequence in a hairpin oligonucleotide (for instance Molecular Beacon-type primer), the tailing of such a hairpin oligonucleotide with a 'universal' tail (for instance Sunrise-type probe and Amplifluor TM -type primer). A special type of hairpin oligonucleotide incorporates in the hairpin a sequence capable of hybridizing to part of the newly amplified target DNA. Amplification of the hairpin is prevented by the incorporation of a blocking nonamplifiable monomer (such as hexethylene glycol). A fluorescent signal is generated after opening of the hairpin due to hybridization of the hairpin loop with the amplified target DNA. This type of hairpin oligonucleotide is known as scorpion primers (Whitcombe et al. 1999, Nat Biotechnol 17:804-807). Another special type of oligonucleotide is a padlock oligonucleotide (or circularizable, open circle, or C-oligonucleotide) that are used in RCA (rolling circle amplification). Such oligonucleotides may also comprise a 3'-terminal mismatching nucleotide and / or, optionally, a 3'-proximal mismatching nucleotide, which can be particularly useful for performing polymorphism-specific PCR and LCR (ligase chain reaction) or any modification of PCR or LCR. Such oligonucleotide may can comprise or consist of at least and / or comprise or consist of up to 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100, 105, 110, 115, 120, 125, 130, 135, 140, 145, 150, 155, 160, 165, 170, 175, 180, 185, 190, 195, 200 or more contiguous nucleotides.

[0565] The analyte nucleic acid, in particular the analyte nucleic acid of a retrotransposon of interest can be any type of nucleic acid, which will be dependent on the manipulation steps (such as isolation and / or purification and / or duplication, multiplication or amplification) applied to the nucleic acid of the gene of interest in the biological sample; as such it can be DNA, RNA, cDNA, may comprise modified nucleotides, or may be hybrids of DNA and / or RNA and / or modified nucleotides, and can be single- or doublestranded or may be a triplex-forming nucleic acid.

[0566] The artificial, man-made, non-naturally occurring oligonucleotide(s) as applied in the above detection methods can be probe(s) or a primer(s), or a combination of both. A probe capable of specifically hybridizing with a target nucleic acid is an oligonucleotide mainly hybridizing to one specific nucleic acid sequence in a mixture of many different nucleic acid sequences. Specific hybridization is meant to result, upon detection of the specifically formed hybrids, in a signal-to- noise ratio (wherein the signal represents specific hybridization and the noise represents unspecific hybridization) sufficiently high to enable unambiguous detection of said specific hybrids. In a specific case specific hybridization allows discrimination of up to a single nucleotide mismatch between the probe and the target nucleic acids. Conditions allowing specific hybridization generally are stringent but can obviously be varied depending on the complexity (size, GC-content, overall identity, etc.) of the probe(s) and / or target nucleic acid molecules. Specificity of a probe in hybridizing with a nucleic acid can be improved by introducing modified nucleotides in said probe.

[0567] A primer capable of directing specific amplification of a target nucleic acid is the at least one oligonucleotide in a nucleic acid amplification reaction mixture that is required to obtain specific amplification of a target nucleic acid. Nucleic acid amplification can be linear or exponential and can result in an amplified single nucleic acid of a single- or double-stranded nucleic acid or can result in both strands of a double-stranded nucleic acid. Specificity of a primer in directing amplification of a nucleic acid can be improved by introducing modified nucleotides in said primer. The fact that a primer does not have to match exactly with the corresponding template or target sequence to warrant specific amplification of said template or target sequence is amply documented in literature (for instance: Kwok et al. 1990, Nucl Acids Res 18:999-1005. Primers as short as 8 nucleotides in length have been applied successfully in directing specific amplification of a target nucleic acid molecule (e.g. Majzoub et al. 1983, J Biol Chem 258:14061-14064).

[0568] A nucleotide is meant to include any naturally occurring nucleotide as well as any modified nucleotide wherein said modification can occur in the structure of the nucleotide base (modification relative to A, T, G, C, or U) and / or in the structure of the nucleotide sugar (modification relative to ribose or deoxyribose). Any of the modifications can be introduced in a nucleic acid or oligonucleotide to increase / decrease stability and / or reactivity of the nucleic acid or oligonucleotide and / or for other purposes such as labelling of the nucleic acid or oligonucleotide. Modified nucleotides include phophorothioates, alkylphophorothioates, methylphosphonate, phosphoramidate, peptide nucleic acid monomers and locked nucleic acid monomers, cyclic nucleotides, and labelled nucleotides (i.e. nucleotides conjugated to a label which can be isotopic (<32>P, <35>S, etc.) or non-isotopic (biotin, digoxigenin, phosphorescent labels, fluorescent labels, fluorescence quenching moiety, etc.)). Other modifications are described higher (see description on oligonucleotides).

[0569] Nucleotide acid amplification is meant to include all methods resulting in multiplication of the number of a target nucleic acid. Nucleotide sequence amplification methods include the polymerase chain reaction (PCR; DNA amplification), strand displacement amplification (SDA; DNA amplification), transcription-based amplification system (TAS; RNA amplification), self-sustained sequence replication (3SR; RNA amplification), nucleic acid sequence-based amplification (NASBA; RNA amplification), transcription-mediated amplification (TMA; RNA amplification), Qbeta-replicase-mediated amplification and run-off transcription. During amplification, the amplified products can be conveniently labeled either using labeled primers or by incorporating labeled nucleotides.

[0570] The most widely spread nucleotide sequence amplification technique is PCR. The target DNA is exponentially amplified. Many methods rely on PCR including AFLP (amplified fragment length polymorphism), IRS-PCR (interspersed repetitive sequence PCR), iPCR (inverse PCR), RAPD (rapid amplification of polymorphic DNA), RT-PCR (reverse transcription PCR) and real-time PCR. RT-PCR can be performed with a single thermostable enzyme having both reverse transcriptase and DNA polymerase activity (Myers et al. 1991, Biochem 30:7661-7666). Alternatively, a single tube-reaction with two enzymes (reverse transcriptase and thermostable DNA polymerase) is possible (Cusi et al. 1994, Biotechniques 17:1034-1036).

[0571] Solid phases, solid matrices or solid supports on which molecules, e.g., nucleic acids, analyte nucleic acids and / or oligonucleotides as described hereinabove, may be bound (or captured, absorbed, adsorbed, linked, coated, immobilized; covalently or non-covalently) comprise beads or the wells or cups of microtiter plates, or may be in other forms, such as solid or hollow rods or pipettes, particles, e.g., from 0.1 pm to 5 mm in diameter (e.g. "latex" particles, protein particles, or any other synthetic or natural particulate material), microspheres or beads (e.g. protein A beads, magnetic beads). A solid phase may be of a plastic or polymeric material such as nitrocellulose, polyvinyl chloride, polystyrene, polyamide, polyvinylidene fluoride or other synthetic polymers. Other solid phases include membranes, sheets, strips, films and coatings of any porous, fibrous or bibulous material such as nylon, polyvinyl chloride or another synthetic polymer, a natural polymer (or a derivative thereof) such as cellulose (or a derivative thereof such as cellulose acetate or nitrocellulose). Fibers or slides of glass, fused silica or quartz are other examples of solid supports. Paper, e.g., diazotized paper may also be applied as solid phase. Clearly, molecules such as nucleic acids, analyte nucleic acids and / or oligonucleotides as described hereinabove, may be bound, captured, absorbed, adsorbed, linked or coated to any solid phase suitable for use in hybridization assay (irrespective of the format, for instance capture assay, reverse hybridization assay, or dynamic allele-specific hybridization (DASH)). Said molecules, such as nucleic acids, analyte nucleic acids and / or oligonucleotides as described hereinabove, can be present on a solid phase in defined zones such as spots or lines. Such solid phases may be incorporated in a component such as a cartridge of e.g. an assay device. Any of the solid phases described above can be developed, e.g. automatically developed in an assay device.

[0572] Quantification of amplified DNA can be performed concurrent with or during the amplification. Techniques include real-time PCR or (semi-)quantitative polymerase chain reaction (qPCR). One common method includes measurement of a non-sequence specific fluorescent dye (e.g. SYBR Green) intercalating in any double-stranded DNA. Quantification of multiple amplicons with different melting points can be followed simultaneously by means of following or analyzing the melting reaction (melting curve analysis or melt curve analysis; which can be performed at high resolution, see, e.g. Wittwer et al. 2003, Clin Chem 843-860; an alternative method is denaturing gel gradient electrophoresis, DGGE; both methods were compared in e.g. Tindall et al. 2009, Hum Mutat 30:857-859).

[0573] Another common method includes measurement of sequence-specific labelled probe bound to its complementary sequence; such probe also carries a quencher and the label is only measurable upon exonucleolytic release from the probe (hydrolysis probes such as TaqMan probes) or upon hybridization with the target sequence (hairpin probes such as molecular beacons which carry an internally quenched fluorophore whose fluorescence is restored upon unfolding the hairpin). This latter method allows for multiplexing by e.g. using mixtures of probes each tagged with a different label e.g. fluorescing at a different wavelength.

[0574] Exciton-controlled hybridization-sensitive fluorescent oligonucleotide (ECHO) probes also allow for multiplexing. The hybridization-sensitive fluorescence emission of ECHO probes and the further modification of probes have made possible multicolor RNA imaging in living cells and facile detection of gene polymorphisms (Okamoto 2011, Chem Soc Rev, 40:5815-5828).

[0575] Other methods of quantifying expression include SAGE (Serial Analysis of Gene Expression) and MPSS (Massively Parallel Signature Sequencing), each involving reverse-transcription of RNA.

[0576] With "assaying" or "determining" or "detecting" and the like (e.g. assessing, measuring) is meant that a biological sample, suspected of comprising a target nucleic acid (such as a nucleic acid of interest as described herein), is processed as to generate a readable signal in case the target nucleic acid is actually present in the biological sample. Such processing may include, as described above, a step of producing an analyte nucleic acid. Simple detection of a produced readable signal indicates the presence of a target or analyte nucleic acid in the biological sample. When in addition the amplitude of the produced readable signal is determined, this allows for quantification of levels of a target or analyte nucleic acid as present in a biological sample.

[0577] In particular, the readable signal may be a signal-to-noise ratio (wherein the signal represents specific detection and the noise represents unspecific detection) of an assay optimized to yield signal-to-noise ratios sufficiently high to enable unambiguous detection and / or quantification of the target nucleic acid. The noise signal, or background signal, can be determined e.g. on biological samples not comprising the target or analyte nucleic acid of interest, e.g. control samples, or comprising the required reference level of the target or analyte nucleic acid of interest, e.g. reference samples. Such noise or background signal may also serve as comparator value for determining an increase or decrease of the level of a target or analyte nucleic acid in the biological sample, e.g. in a biological sample taken from a subject suffering from a disease or disorder, further e.g. before start of a treatment and during treatment.

[0578] The readable signal may be produced with all required components in solution or may be produced with some of the required components in solution and some bound to a solid support. Said signals include, e.g., fluorescent signals, (chemi)luminescent signals, phosphorescence signals, radiation signals, light or color signals, optical density signals, hybridization signals, mass spectrometric signals, spectrometric signals, chromatographic signals, electric signals, electronic signals, electrophoretic signals, real-time PCR signals, PCR signals, LCR signals, Invader-assay signals, sequencing signals (by any method such as Sanger dideoxy sequencing, pyrosequencing, 454 sequencing, single-base extension sequencing, sequencing by ligation, sequencing by synthesis, "next-generation" sequencing (NGS)(van Dijk et al. 2014, Trends Genet 30:418-426)), melting curve signals etc. An assay may be run automatically or semi- automatically in an assay device. In view of its relatively low costs compared to e.g. very costly cancer therapies, NGS is finding its way to routine clinical care (Ratner 2018, Nature Biotechnol 36:484).

[0579] Specific hybridization of an oligonucleotide (whether or not comprising one or more modified nucleotides) to its target sequence is to be understood to occur under stringent conditions as generally known in the art (e.g. Sambrook et al. 1989. Molecular Cloning. A laboratory manual. CSHL Press). However, depending to the hybridization solution (SSC, SSPE, etc.), oligonucleotides should be hybridized at their appropriate temperature in order to attain sufficient specificity. In order to allow hybridization to occur, the target nucleic acid molecules are generally thermally, chemically (e.g. by NaOH) or electrochemically denatured to melt a double strand into two single strands and / or to remove hairpins or other secondary structures from single stranded nucleic acids. The stringency of hybridization is influenced by conditions such as temperature, salt concentration and hybridization buffer composition. High stringency conditions for hybridization include high temperature and / or low salt concentration (salts include NaCI and Na3-citrate) and / or the inclusion of formamide in the hybridization buffer and / or lowering the concentration of compounds such as SDS (detergent) in the hybridization buffer and / or exclusion of compounds such as dextran sulfate or polyethylene glycol (promoting molecular crowding) from the hybridization buffer. Conventional hybridization conditions are described in e.g. Sambrook et al. 1989 (Molecular Cloning. A laboratory manual. CSHL Press) but the skilled craftsman will appreciate that numerous different hybridization conditions can be designed in function of the known or the expected homology and / or length of the nucleic acid sequence. Generally, for hybridizations with DNA oligonucleotides without formamide, a temperature of 68 DEG C, and for hybridization with formamide, 50% (v / v), a temperature of 42 DEG C is recommended. For hybridizations with oligonucleotides, the optimal conditions (formamide concentration and / or temperature) depend on the length and base composition of the probe and must be determined individually. In general, optimal hybridization for oligonucleotides of about 10 to 50 bases in length occurs approximately 5 DEG C below the melting temperature for a given duplex. Incubation at temperatures below the optimum may allow mismatched sequences to hybridize and can therefor result in reduced specificity. When using RNA oligonucleotides with formamide (50% v / v) it is recommend to use a hybridization temperature of 68 DEG C for detection of target RNA and of 50 DEG C for detection of target DNA. Alternatively, a high SDS hybridization solution can be utilized (Church et al. 1984, Proc Natl Acad Sci USA 81:1991-1995). The specificity of hybridization can furthermore be ensured through the presence of a crosslinking moiety on the oligonucleotide (e.g. Huan et al. 2000, Biotechniques 28: 254-255; WOOO / 14281). Said crosslinking moiety enables covalent linking of the oligonucleotide with the target nucleotide sequence and hence allows stringent washing conditions. Such a crosslinking oligonucleotide can furthermore comprise another label suitable for detection / quantification of the oligonucleotide hybridized to the target.

[0580] Kits

[0581] In view of all above, kits, such as diagnostic or prognostic kits or kits of parts, can be designed that are tailored to enable any of:

[0582] - a method of separating / distinguishing AD patients or subjects at risk of developing AD from healthy / control subjects, or a method of diagnosing / determining / assessing / prognosticating / identifying a subject as AD patient or as potential AD patient, or as subject at risk of developing AD, as described hereinabove;

[0583] - a method of screening for AD drugs / medicaments as describe hereinabove;

[0584] - a method of selecting a subject for inclusion in a clinical trial of a candidate AD drug / medicament as described hereinabove; or

[0585] - a method of selecting a subject for treatment with a drug / medicament as described hereinabove.

[0586] Such kits in particular comprise the tools, such as oligonucleotides, to detect the SGVs as described herein (single SGVs as occurring in the (pathogenic) SGV combinations are listed in Table 1). The oligonucleotides can be primers and / or probes, such as allele-specific primers and / or probes. In particular, the primers and / or probes are labeled primers and / or probes; primers and / or probes comprising non-naturally occurring nucleotides; hairpin or structurally locked primers and / or probes; or a combination thereof. In particular, the oligonucleotides are capable of detecting the status of an envisaged SGV. In particular, the oligonucleotides comprise a sequence specifically hybridizing to said SGV or in the immediate vicinity of said SGV. In particular, the oligonucleotide is comprising least one modified or non-naturally occurring nucleotide (as described hereinabove). Further in particular, the oligonucleotide may be part of a primer and probe set, of which set at least one primer or probe is comprising a sequence specifically hybridizing to the envisaged SGV or in the immediate vicinity of said SGV. Such kits / diagnostic kits can alternatively comprise a multi-membered set of oligonucleotides, wherein each member of the set comprises at least one modified or non-naturally occurring nucleotide and a sequence specifically hybridizing to one of the envisaged SGVs or in the immediate vicinity of said SGV. Such kits / diagnostic kits can alternatively comprise a plurality of separate primer and probe sets, wherein each set is comprising a primer or probe comprising of which at least one of the primer or probe is comprising a modified or non-naturally occurring nucleotide, and wherein each set comprises a primer or probe of which at least one of the primer or probe is comprising a sequence specifically hybridizing to one of the envisaged SGVs or in the immediate vicinity of said SGV. A non-naturally occurring nucleotide may be a nucleotide that is chemically different from a nucleotide present in a living cell (such as a labelled nucleotide), or may be a chemically naturally occurring nucleotide but which is mutated relative to the natural target nucleic acid on which the oligonucleotide is specifically hybridizing.

[0587] Depending on the method of detection of the SGVs as described herein, such kits may further comprise reagents such as reagents required for extraction of DNA from cells, reagents for amplification of DNA, hybridization reagents, DNA intercalating dyes, reaction vials, and a kit instruction manual or a kit manual.

[0588] In one specific embodiment, such kits / diagnostic kits are including the tools for detecting the status of, in total, at most 250 SGV combinations as described herein, or at most 225, 200, 175, 150, 125, 111, 110, 105, 100, 95, 90, 85, 80, 79, 78, 77, 76, 75, 74, 73, 72, 71, 70, 69, 68, 67, 66, 65, 64, 63, 62, 61, 60, 59, 58, 57, 56, 55, 54, 53, 52, 51, 50, 49, 48, 47, 46, 45, 44, 43, 42, 41, 40, 39, 38, 37, 36, 35, 34, 33, 32, 31, 30, 29, 28, 27, 26, 25, 24, 23, 22, 21, 20 ,19, 18, 17, 16, 15, 14, 13, 12, 11, 10, 9 , 8, 7, 6, 5, 4, 3, 2 SGV combinations as described herein, or at most 1 SGV combination as described herein.

[0589] In a further specific embodiment, such kits / diagnostic kit contain the components and / or reagents enabling multiplex analysis of more than SGV combination.

[0590] Computer / computer system

[0591] A further aspect of the current disclosure relates to computer systems or machine readable media comprising the SGV combination selection algorithm trained to select SGV combinations for separating AD patients or subjects at risk of developing AD from healthy subjects, wherein the SGV combination selection algorithm was trained based on the input features of a database comprising a set of data of AD patients and a set of data of healthy subjects. More in particular, such computer systems or machine readable media are comprising the computer implemented SGV combination selection algorithm as described hereinabove.

[0592] A computer or computer system as mentioned herein may utilize one or more subsystems. A computer or computer system may be a single computer apparatus comprising the one or more subsystems (e.g. internal components), or may be multiple computers or multiple computer apparatuses each being a subsystem, and optionally, each comprising one or more own subsystems. Desktops, laptops, mainframe servers, tablets, mobile phones etc. all are computers or computer systems. The subsystems are usually interconnected and include a (central) processor (single-core processor, multi-core processor on a same integrated chip, or multiple processing units on a single circuit board or networked) capable of executing instructions, an input / output (I / O) controller, and a storage device (external, internal, peripheral, cloud, any medium readable by a computer or computer system). Input devices include keyboards, scanners, a computer mouse, camera, microphone, etc. In particular, the input device is a data collection (e.g. relevant data of the European Alzheimer's Disease Biobank (EAD B) ) or data generating device (which by itself may comprise a computer or computer system), such as a polynucleotide sequencing device (whether automated or not). Collected or generated data are fed to a computer or computer system designed to analyze the collected or generated data; this may be an ordinary computer system on which data analyzing software is installed (on a storage device) or which is capable of accessing data analyzing software (e.g. installed in or transmitted from a network) and whereby the processor of the computer system is instructed by the data analysis software on how to process the collected or generated data fed to the computer system, and how to display these via a display adapter to an output device. Output devices are further subsystems and comprise printers, monitors, computer readable medium. Input and output devices are usually connected to a computer or computer system via input / output ports to one another or via a network.

[0593] The specific combination of hardware and software allows implementation of e.g. analysis of data generated by a polynucleotide sequencing device, or of e.g. determination of the pathogenicity of a single SGV or of a SGV combination, such as by a trained or machine learned data classification algorithm. Different software packages (proprietary or open source) can be run on a computer or computer system to achieve the desired degree of data analysis. Output of one computerized data analysis can be the input of a subsequent computerized data analysis step, hence creating an analysis pipeline. Software components can be written in different codes (e.g. Java, C, C++, Perl, Python) as long as the computer processor is able to execute the functions of the software component.

[0594] The methods of the invention may be computer-implemented methods, or methods that are assisted or supported (in part) by a computer or by a computer system. For instance, information reflecting the EADB can be provided in user readable format by at least one / another processor. The same or a further processor may be calculating, as outlined herein, the pathogenicity of a single test SGV or of a test SGV combination (such as relative to a control, standard or reference) from the information received. The one or more processors may be coupled to random access memory operating under control of or in conjunction with a computer operating system. The processors may be included in one or more servers, clusters, or other computers or hardware resources, or may be implemented using cloud-based resources. The operating system may be, for example, a distribution of the LinuxTM operating system, the UnixTM operating system, or other open- source or proprietary operating system or platform. Processors may communicate with data storage devices, such as a database stored on a hard drive or drive array or such as a computer or machine readable medium, to access or store program instructions other data. Processors may further communicate via a network interface, which in turn may communicate via the one or more networks, such as the Internet or other public or private networks, such that a query or other request may be received from a client, or other device or service. Such computer-implemented methods (or such methods that are assisted or supported by a computer) may be provided as a kit or as part of a kit. The bioinformatics software required to perform (part of) the computer-implemented methods, i.e. a computer program product, may also be part of a kit, or may be provided as an individual product. A computer product may also consist of a computer or machine readable medium (in any form such as disks (hard, soft disks etc.), cards (memory cards etc.), tapes, sticks (memory, USB sticks etc.), microchips, DVDs, CDs, etc.) which is storing any of the instructions, computer program, or bioinformatics software enabling a computer system to perform at least one of the analysis of the herein described methods and / or to perform at least one calculation as described herein. Furthermore, the computer or computer system can be set up or configurated such that it is compliant with GDPR.

[0595] EXAMPLES

[0596] EXAMPLE 1. Combinations of short genetic variants (SGVs) for separating Alzheimer disease (AD) patients (or subjects at risk of developing AD) and healthy subjects.

[0597] An unbiased analysis searching for combinations of the 83 Alzheimer disease (AD) genome-wide association study (GWAS) significant short genetic variants (SGVs) that separate AD patients from controls with high sensitivity (Table 1; Bellenguez et al. 2022, Nature Genetics 54:412-436) was performed. The same SGVs are annotated based on the GRCh / hg38 reference genome (Table 1, last column). The majority of SGVs are single nucleotide polymorphisms (SGVs) except for 4 SGVs characterized by short deletions, i.e. the SGVs in lines ("#") 32, 33, 41, and 59 of Table 1. In view of these SGVs separating AD patients from controls, these SGVs are also termed "risk alleles", "AD risk alleles", and are interchangeably referred to herein sometimes simply as "variants" or "polymorphisms" (to be understood in the context of AD). Table 1 does not include the risk alleles APOE rs429358 and APOE rs7412 (APOE rs429358: chrl9 position 44908684 T C; APOE rs7412: chrl9 position 44908822 C T in GRCh / hg38). Table 1 further included are the odds ratio's (OR) with 95% confidence intervals (OR 95 Cl) of the effect in AD development of the minor alleles, as reported by Bellenguez et al. 2022. The European Alzheimer's Disease Biobank (EADB) cohort was used as starting basis for the analysis. The EADB consists of >20.000 clinically diagnosed AD patients and >21.000 unrelated age-matched controls. This dataset has been put together thanks to a long-term collaboration between 15 different European countries.

[0598] First, the EADB was adapted by substituting the genotype data from each position with the number of risk alleles (RA) present at each of the 85 variants (including the 2 APOE SGVs). The RA was defined as the allele having an odd's ratio (OR) >1 as determined by Bellenguez et al. 2022 (Nature Genetics 54:412- 436), therefore for RA=0 no risk alleles are found in a specific position, whereas for RA=2 both alleles present a variant associated with increased risk for AD.

[0599] As a remark, there are zero samples with the very rare APOE e3r isoform, thus in this dataset, 2 RAs at APOE rs429358 imply the presence of 2 RAs at APOE rs7412, together conforming APOE4 / 4.

[0600] Subsequently, using the Caret R package, the dataset was split into two different training groups ("EADB train 1" and "EADB train 2") containing each of them 45% of the total individuals from EADB dataset. The distribution of both numerical and categorical variables was constant when compared to the original EADB dataset. The remaining 10% of the EADB dataset ("EADB test") was reserved to assess the accuracy of identified combinations in diagnosing AD.

[0601] Independently, for purposes of an alternative analysis, the three datasets ("EADB train 1", "EADB train 2" and "EADB test") were split based on the number of APOE4 alleles. Due to the absence of APOE e3r carriers, this is determined by the number of risk alleles in APOE rs429358, meaning that the APOE rs429358 - / - subgroup contain 0 APOE4 alleles, the APOE rs429358 + / - subgroup contains 1 APOE4 allele, and the APOE rs429358 + / + subgroup contains 2 APOE4 alleles.

[0602] Next, multifactor dimensionality reduction (MDR) analysis in the JAVA software MDR.v3.0.2 (Ritchie et al. 2001, Am J Human Genetics 69:138-147) was performed. This analysis was run independently on both EADB train 1 and EADB train 2, and independently in each of the APOE4 subgroups, with the following common settings: Cross-Validation Count=10, Top Models Tracked=3000. Further, 10 different random seeds were set to avoid the arbitrarily organization of the dataset.

[0603] In a next step, combinations of 2, 3, and 4 variants / polymorphisms / SGVs, both in the absence and in the presence of APOE polymorphism (as represented by rs429358 and rs7412 genotypes)were selected in the analysis not including prior stratification of the datasets according to APOE4 background. Therefore, combinations that were not replicated by MDR in both EADB train 1 and EADB train 2 datasets (are not common in both EADB train 1 and EADB train 2 datasets) were filtered out. As such about ~35% of the combos containing an APOE4 risk allele and ~65% of the combos without APOE4 risk allele were removed. An independent similar analysis was performed to select combinations of 2 SGVs from the 3 subgroups stratified according to their APOE4 background. Subsequently, in order to assess the pathogenicity of the combinations, a logistic regression ( LR) analysis was run. Covariates to be introduced in the glm (generalized linear model) model were selected by running a likelihood-ratio test with the base R function Irtest (likelihood ratio test of nested models), which yielded the following confounding factors / covariates: sex, age, APOE status and 20 genetic principal components (gPCs).

[0604] Genetic principal components (gPCs) are a way of summarizing the genetic variations across a large number of genetic markers in a population (in the present case, the whole genome of the individuals that are part from the EADB dataset). They are derived from a technique called Principal Component Analysis (PCA). PCA is applied to a set of genetic markers (whole genomes in the present case) in order to identify the underlying genetic structure of the population by finding the principal components that explain most of the variation in the data. This is a common way of identifying genetic variation that can be structured along different geographic and ancestry lines, which could lead to spurious associations in genetic studies (false positive and false negative results). More specifically, if a dataset is composed by individuals from different countries (the individuals from this study are from 15 different European countries) there may be differences in the genetic background that may be due to a different ancestry origin, rather than the trait (Alzheimer's Disease or control) being studied. Once identified, these gPCs are used as covariates in the LR model, so that pathogenic combinations of AD variants are identified those, independently of the ancestry origin. Those gPCs are stored in a principal component matrix, which contains one given value for each of the individuals, with a total of 20 different gPCs / individual in the current case.

[0605] Combinations that deviate from the expected additive effect (additive effect: sum of single effects of the single SGVs) were determined by introducing the main effect of the variants as covariates. As additive effects are filtered out, synergistic effects are retained (synergistic effect: greater than sum of single effects of the single SGVs). In total, LR was run on 5883 combinations of variants / polymorphisms / SGVs containing / including APOE and on 3084 combinations of variants / polymorphisms / SGVs without / excluding APOE. Only combinations with p<0.05 and beta > 0 were retained.

[0606] Combinations that are significantly (p value < 0.05) and synergistically (odds ratio or OR >1) pathogenic were subsequently checked with the EADB test dataset. Table 2 lists the number of identified synergistic combinations of 2, 3, and 4 SGVs. Table 2 indicates in which further Tables the specific combinations and their properties are listed.

[0607] With respect to the combos of 2 SGVs including APOE4, the analysis was pushed to keep searching for combinations not containing APOE. Due to this, numerous SGV pairs from the combo's of 2 SGVs including APOE4 and from the combo's of 2 SGVs excluding SGV overlap (compare e.g. Tables 3A and 4A). Tables 3A, 5A, and 7A refer to combinations (pairs / triplets / quartets) of 83 variants (including APOE polymorphism) that were A) replicated by MDR in train set 1 and train set 2, B) significant (p<0,05) and pathogenic (beta > 0) in the logistic regression model, containing the following information: SGV# as the SGV# id from the variant # in the combination, combining locus name and rsID - wherein the "#" can be 1, 2, 3 or 4; RA SGV# as the number of risk alleles in SGV#; Total freq as the frequency of individuals from the full EADB dataset containing the given combinatorial genotype; Freq AD as the percentage of individuals containing the given combinatorial genotype that are AD patients; Freq C as the percentage of individuals containing the given combinatorial genotype that are Controls.

[0608] Tables 3B, 5B, and 7B repeat in part the information of Tables 3A, 5A, and 7A (SGV#, RA SGV#) and further include for each listed combination the total RA (sum of the respective RA SGV#'s), the odds ratio (OR), the 95% confidence interval of the OR (lower limit "OR L95"; and upper limit "OR U95") and the probability (p-value "P regr"). Whereas the SGV combinations in Tables 3A, 5A, and 7A on the one hand, and in Tables 3B, 5B, and 7B on the other hand is the same, the order of the combinations is different. Tables 3B, 5B, and 7B includes the values of logistic regression analysis including the main effect of the single SGVs as covariates. Table 3B, by way of example, further includes the values of logistic regression analysis NOT including the main effect of the single SGVs as covariates.

[0609] Tables 4A, 6A, and 8A refer to combinations (pairs / triplets / quartets) of 83 variants (excluding the APOE polymorphism) that were A) replicated by MDR in train set 1 and train set 2, B) significant (p<0,05) and pathogenic (beta > 0) in the logistic regression model, containing the following information: SGV# as the SGV# id from the variant # in the combination, combining locus name and rsID - wherein the "#" can be 1, 2, 3 or 4; RA SGV# as the number of risk alleles in SGV#; Total freq as the frequency of individuals from the full EADB dataset containing the given combinatorial genotype; Freq AD as the percentage of individuals containing the given combinatorial genotype that are AD patients; Freq C as the percentage of individuals containing the given combinatorial genotype that are Controls.

[0610] Tables 4B, 6B, and 8B repeat in part the information of Tables 4A, 6A, and 8A (SGV#, RA SGV#) and further include for each listed combination the total RA (sum of the respective RA SGV#'s), the odds ratio (OR), the 95% confidence interval of the OR (lower limit "OR L95"; and upper limit "OR U95") and the probability (p-value "P regr"). Whereas the SGV combinations in Tables 4A, 6A, and 8A on the one hand, and in Tables 4B, 6B, and 8B on the other hand is the same, the order of the combinations is different.

[0611] Tables 3C, 4C, 5C, 6C, 7C, and 8C summarize the frequency of occurrence of the listed individual SGVs in the total number of combinations of Tables 3A / B, 4A / B, 5A / B, 6A / B, 7A / B, and 8A / B, respectively. Importantly, the odds ratio's as reported in Tables 3B, 4B, 5B, 6B, 7B, and 8B specifically relate to the synergistic pathogenic effect of the respective SGV combinations. These ORs therefore cannot be compared head-to-head with the ORs reported for the single SGVs (reported by Bellenguez et al. 2022, Nature Genetics 54:412-436; included in Table 1). More in particular, the ORs reported for the single SGVs refer to the minor allele of each SGV, with the minor allele not always being the risk allele.

[0612] Tables 9A, 10A, and 11A refer to combinations (pairs) of 83 variants (excluding the APOE polymorphism) that were A) replicated by MDR in train set 1 and train set 2 per subgroup stratified according to APOE4 background, B) significant (p<0,05) and pathogenic (beta > 0) in the logistic regression model, containing the following information: SGV1 is the SGV1 id from the variant 1 in the combination, combining locus name and rsID; SGV2 is the SGV2 id from the variant 2 in the combination; RA SGV1 is the number of risk alleles in SGV1; RA SGV2 is the number of risk alleles in SGV2. Further included for each listed combination are the total RA (sum of the respective RA SGV#'s), the odds ratio (OR), the 95% confidence interval of the OR (lower limit "OR L95"; and upper limit "OR U95") and the probability (p-value "P regr") of the logistic regression analysis including the main effect of the single SGVs as covariate.

[0613] Many regions of the genome that are inherited together. The likelihood for two genetic variants, such as the individual pathogenic SGVs as described herein ("AD-SGV", see Table 1 which does not include the two APOEE4 pathogenic SGVs) and another SGV to be inherited together is measured by Linkage Disequilibrium (LD) analysis, which basically provides a score between 0 (independent loci, inherited separately) and 1 (dependent loci, inherited together). LD was calculated in plink 1.9 (cog- genomics.org / plink / 1.9 / ) for each of the individual AD-SGVs, yielding a list of SGVs in LD ("LD-SGVs") with the AD-SGVs. This LD analysis was run on a 1000 Genomes panel for European non-Finnish populations as a reference panel. In addition, LD-SGVs in the list of Table 12 are further restricted to those (i) with an LD score equal to or higher than 0.8 and (ii) within a distance of 1000 kilobases from an AD-SGV.

[0614] All the combinations, independently of their APOE-stratification category, were evaluated by ROC analysis to test whether the synergistic pathogenicity that was predicted in the training datasets was increasing the prediction accuracy of Alzheimer's Disease patients when compared to the standard additive model composed by the single effects of the 83 lead variants (excluding APOE). 151 combinations (51 combinations from APOE- / - subgroup, 48 combinations from the APOE+ / - subgroup, and 42 combinations from the APOE + / + subgroup) incremented, individually, the overall prediction accuracy of the additive effects of the 83 pathogenic variants (Table 1; excluding APOE). To test whether the combined effect of these pre-selected combinations increase the prediction accuracy of Alzheimer's Disease patients when compare to their single additive effects, we combined the main effects of the 83 different pathogenic variants (excluding APOE) and the pathogenic synergistic effects of these 151 preselected combinations in a model called combinatorial Genetic Risk Score (cGRS), and compare its performance with a model composed only with the main effects of the variants, named additive Genetic Risk Score (aG RS). Overall, the cGRS model (AUC = 0.76) significantly outperforms the aGRS model (AUC = 0.71). This is also true when splitting by APOE status. In APOE - / - carriers, the cGRS model (0.75) outperforms the aGRS model (AUC = 0.73). In APOE+ / - carriers, the cGRS model (AUC = 0.76) outperforms the aGRS model (AUC = 0.74). In APOE + / + carriers, the cGRS model (AUC = 0.89) outperforms the aGRS model (AUC = 0.78). All AUC differences were significant when evaluated by bootstrap method.

[0615] TABLES

[0616] Table 1. Individual SGVs as basis for combinations. Information for all the 83 independent variants (excluding the APOE polymorphism) from Bellenguez et al. 2022 (Nature Genetics 54:412-436), including "rs ID" as the rs code assigned to the given variants (according to short genetic variants database release 153 / dbSGV build 153); "locus" as the locus name of the variant; "Effect Allele" as the allele analysed in Bellenguez et al. 2022 (Nature Genetics 54:412-436); "Other allele" as the other existing allele within the same position; "Effect Allele Freq" as the allele frequency of the effect allele in the all the individuals employed in Bellenguez et al., 2022 (stage I and stage II); "Beta coef" Beta coefficient as the estimate calculated in Bellenguez et al. 2022 for the effect allele, corrected per gPCs x20 (20 genetic principal components, see further) and genotyping centre, when necessary; "RA": risk allele or pathogenic allele. Genomic position of the individual SGVs are further annotated based on the GRCh / hg38 reference genome, and in the format chrZ XXXXX A B, meaning chromosome number (Z), chromosome position (XXXXX), reference allele ("other allele") in the genome (A), and effect allele (B). Further included are the odds ratio's (OR) with 95% confidence intervals (OR 95 Cl) of the effect in AD development of the minor alleles, as reported by Bellenguez et al. 2022.

[0617]

[0618]

[0619]

[0620] Table 2. Number of synergistic combinations of 2, 3, and 4 SGVs, further referring to the Tables listing the specific combinations and their properties.

[0621] * including AP0E4: meaning that the combinations can, but must not, include an APOE AD risk allele; ** excluding APOE4: meaning that the combinations do not include an APOE AD risk allele

[0622] # dataset stratified in subgroups of individuals comprising 0 (- / -}, 1(+ / -) or 2(+ / +) alleles of APOEe4.

[0623] Table 3A. Combinations of 2 SGVs including APOE4. Combinations (pairs) of 85 variants (including APOE polymorphism) that were A) replicated by MDR in train set 1 and train set 2, B) significant (p<0,05) and pathogenic (beta > 0) in the logistic regression model, containing the following information: SGV1 as the SGV1 id from the variant 1 in the combination, combining locus name and rsID; SGV2 as the SGV2 id from the variant 2 in the combination; RA SGV1 as the number of risk alleles in SGV1; RA SGV2 as the number of risk alleles in SGV2; Total freq as the frequency of individuals from the full EADB dataset containing the given combinatorial genotype; Freq AD as the percentage of individuals containing the given combinatorial genotype that are AD patients; Freq C as the percentage of individuals containing the given combinatorial genotype that are Controls.

[0624] Table 3B. Odds ratio's (OR) of combinations of 2 SGVs including APOE4. This table in part repeats the information of Table 3A (SGV#, RA SGV#) and further includes for each listed combination the total RA (sum of the respective RA SGV#'s), the odds ratio (OR), the 95% confidence interval of the OR (lower limit "OR L95"; and upper limit "OR U95") and the probability (p-value "P regr") of the logistic regression analysis including the main effect of the single SGVs as covariate. "OR1", "OR1 L95", "OR1 U95", and "P regrl" are the values of logistic regression analysis NOT including the main effect of the single SGVs as covariates. Whereas the SGV combinations in Table 3A and in this Table are the same, the order of the combinations is different.

[0625] Table 3C. Frequency of occurrence of single SGV in the combinations of 2 SGVs including APO E4 (as listed in Tables 3A / B).

[0626] Table 4A. Combinations of 2 SGVs excluding APOE4. Combinations (pairs) of 83 variants (including APOE polymorphism) that were A) replicated by MDR in train set 1 and train set 2, B) significant (p<0,05) and pathogenic (beta > 0) in the logistic regression model, containing the following information: SGV1 as the SGV1 id from the variant 1 in the combination, combining locus name and rsID; SGV2 as the SGV2 id from the variant 2 in the combination; RA SGV1 as the number of risk alleles in SGV1; RA SGV2 as the number of risk alleles in SGV2; Total freq as the frequency of individuals from the full EADB dataset containing the given combinatorial genotype; Freq AD as the percentage of individuals containing the given combinatorial genotype that are AD patients; Freq C as the percentage of individuals containing the given combinatorial genotype that are Controls.

[0627] Table 4B. Odds ratio's (OR) of combinations of 2 SGVs excluding APOE4. This table in part repeats the information of Table 4A (SGV#, RA SGV#) and further includes for each listed combination the total RA (sum of the respective RA SGV#'s), the odds ratio (OR), the 95% confidence interval of the OR (lower limit "OR L95"; and upper limit "OR U95") and the probability (p-value "P regr") of the logistic regression analysis including the main effect of the single SGVs as covariate. Whereas the SGV combinations in Table 4A and in this Table are the same, the order of the combinations is different. Table 4C. Frequency of occurrence of single SGV in the combinations of 2 SGVs excluding APOE4 (as listed in Tables 4A / B). Table 5A. Combinations of 3 SGVs including APOE4. Combinations (trios) of 85 variants that were A) replicated by MDR in train set 1 and train set 2, B) significant (p<0,05) and pathogenic (beta > 0) in the logistic regression model, containing the following information: SGV1 as the SGV1 id from the variant 1 in the combination, combining locus name and rsID; SGV2 as the SGV2 id from the variant 2 in the combination; SGV3 as the SGV3 id from the variant 3 in the combination; RA SGV1 as the number of risk alleles in SGV1; RA SGV2 as the number of risk alleles in SGV2; RA SGV3 as the number of risk alleles in SGV3; Total freq as the frequency of individuals from the full EADB dataset containing the given combinatorial genotype; Total freq as the frequency of individuals from the full EADB dataset containing the given combinatorial genotype; Freq AD as the percentage of individuals containing the given combinatorial genotype that are AD patients; Freq C as the percentage of individuals containing the given combinatorial genotype that are Controls. Table 5B. Odds ratio's (OR) of combinations of 3 SGVs including APOE4. This table in part repeats the information of Table 5A (SGV#, RA SGV#) and further includes for each listed combination the total RA (sum of the respective RA SGV#'s), the odds ratio (OR), the 95% confidence interval of the OR (lower limit "OR L95"; and upper limit "OR U95") and the probability (p-value "P regr") of the logistic regression analysis including the main effect of the single SGVs as covariate. Whereas the SGV combinations in Table 5A and in this Table are the same, the order of the combinations is different.

[0628] Table 5C. Frequency of occurrence of single SGV in the combinations of 3 SGVs including APO E4 (as listed in Tables 5A / B).

[0629] Table 6A. Combinations of 3 SGVs excluding APOE4. Combinations (trios) of 83 variants (including APOE polymorphism) that were A) replicated by MDR in train set 1 and train set 2, B) significant (p<0,05) and pathogenic (beta > 0) in the logistic regression model, containing the following information: SGV1 as the SGV1 id from the variant 1 in the combination, combining locus name and rsID; SGV2 as the SGV2 id from the variant 2 in the combination; SGV3 as the SGV3 id from the variant 3 in the combination; RA SGV1 as the number of risk alleles in SGV1; RA SGV2 as the number of risk alleles in SGV2; RA SGV3 as the number of risk alleles in SGV3; Total freq as the frequency of individuals from the full EADB dataset containing the given combinatorial genotype; Freq AD as the percentage of individuals containing the given combinatorial genotype that are AD patients; Freq C as the percentage of individuals containing the given combinatorial genotype that are Controls.

[0630]

[0631]

[0632]

[0633] Table 6B. Odds ratio's (OR) of combinations of 3 SGVs excluding APOE4. This table in part repeats the information of Table 6A (SGV#, RA SGV#) and further includes for each listed combination the total RA (sum of the respective RA SGV#'s), the odds ratio (OR), the 95% confidence interval of the OR (lower limit "OR L95"; and upper limit "OR U95") and the probability (p-value "P regr") of the logistic regression analysis including the main effect of the single SGVs as covariate. Whereas the SGV combinations in Table 6A and in this Table are the same, the order of the combinations is different.

[0634] Table 6C. Frequency of occurrence of single SGV in the combinations of 3 SGVs excluding APOE4 (as listed in Tables 6A / B).

[0635]

[0636] Table 7A. Combinations of 4 SGVs including APOE4. Combinations (quartets) of 85 variants that were A) replicated by MDR in train set 1 and train set 2, B) significant (p<0,05) and pathogenic (beta > 0) in the logistic regression model, containing the following information: SGV1 as the SGV1 id from the variant 1 in the combination, combining locus name and rsID; SGV2 as the SGV2 id from the variant 2 in the combination; SGV3 as the SGV3 id from the variant 3 in the combination; SGV4 as the SGV4 id from the variant 4 in the combination; RA SGV1 as the number of risk alleles in SGV1; RA SGV2 as the number of risk alleles in SGV2; RA SGV3 as the number of risk alleles in SGV3; RA SGV4 as the number of risk alleles in SGV4; Total freq as the frequency of individuals from the full EADB dataset containing the given combinatorial genotype; Freq AD as the percentage of individuals containing the given combinatorial genotype that are AD patients; Freq C as the percentage of individuals containing the given combinatorial genotype that are Controls.

[0637] Table 7B. Odds ratio's (OR) of combinations of 4 SGVs including APOE4. This table in part repeats the information of Table7A (SGV#, RA SGV#) and further includes for each listed combination the total RA (sum of the respective RA SGV#'s), the odds ratio (OR), the 95% confidence interval of the OR (lower limit "OR L95"; and upper limit "OR U95") and the probability (p-value "P regr") of the logistic regression analysis including the main effect of the single SGVs as covariate. Whereas the SGV combinations in Table 7A and in this Table are the same, the order of the combinations is different.

[0638]

[0639]

[0640] Table 7C. Frequency of occurrence of single SGV in the combinations of 4 SGVs including APOE4 (as listed in Tables 7A / B).

[0641] Table 8A. Combinations of 4 SGVs excluding APOE4. Combinations (quartets) of 83 variants (including APOE polymorphism) that were A) replicated by MDR in train set 1 and train set 2, B) significant (p<0,05) and pathogenic (beta > 0) in the logistic regression model, containing the following information: combo as the combination name; SGV1 as the SGV1 id from the variant 1 in the combination, combining locus name and rsID; SGV2 as the SGV2 id from the variant 2 in the combination; SGV3 as the SGV3 id from the variant 3 in the combination; SGV4 as the SGV4 id from the variant 4 in the combination; RA SGV1 as the number of risk alleles in SGV1; RA SGV2 as the number of risk alleles in SGV2; RA SGV3 as the number of risk alleles in SGV3; RA SGV4 as the number of risk alleles in SGV4; Total freq as the frequency of individuals from the full EADB dataset containing the given combinatorial genotype; Freq AD as the percentage of individuals containing the given combinatorial genotype that are AD patients; Freq C as the percentage of individuals containing the given combinatorial genotype that are Controls.

[0642]

[0643]

[0644]

[0645]

[0646] Table 8B. Odds ratio's (OR) of combinations of 4 SGVs excluding APOE4. This table in part repeats the information of Table 8A (SGV#, RA SGV#) and further includes for each listed combination the total RA (sum of the respective RA SGV#'s), the odds ratio (OR), the 95% confidence interval of the OR (lower limit "OR L95"; and upper limit "OR U95") and the probability (p-value "P regr") of the logistic regression analysis including the main effect of the single SGVs as covariate. Whereas the SGV combinations in Table 8A and in this Table are the same, the order of the combinations is different.

[0647]

[0648]

[0649]

[0650]

[0651]

[0652]

[0653]

[0654]

[0655] Table 8C. Frequency of occurrence of single SGV in the combinations of 4 SGVs excluding APOE4 (as listed in Tables 8A / B).

[0656] Table 9A. Combinations of 2 SGVs in subgroup of individuals comprising 0 (- / -) alleles of APOEe4

[0657] Odds ratio's (OR) of combinations of 2 SGVs selected from the APOEe4 - / - subgroup after stratification of the EADB dataset into 3 subgroups (APOEe4 - / -; APOEe4 + / -; APOEe4 + / +). SGV1 is the SGV1 id from the variant 1 in the combination, combining locus name and rsID; SGV2 is the SGV2 id from the variant 2 in the combination; RA SGV1 is the number of risk alleles in SGV1; RA SGV2 is the number of risk alleles in SGV2. Further included for each listed combination are the total RA (sum of the respective RA SGV#'s), the odds ratio (OR), the 95% confidence interval of the OR (lower limit "OR L95"; and upper limit "OR U95") and the probability (p-value "P regr") of the logistic regression analysis including the main effect of the single SGVs as covariate.

[0658]

[0659] Table 10A. Combinations of 2 SGVs in subgroup of individuals comprising 1(+ / -) allele of APOEe4

[0660] Odds ratio's (OR) of combinations of 2 SGVs selected from the APOEe4 + / - subgroup after stratification of the EADB dataset into 3 subgroups (APOEe4 - / -; APOEe4 + / -; APOEe4 + / +). SGV1 is the SGV1 id from the variant 1 in the combination, combining locus name and rsID; SGV2 is the SGV2 id from the variant 2 in the combination; RA SGV1 is the number of risk alleles in SGV1; RA SGV2 is the number of risk alleles in SGV2. Further included for each listed combination are the total RA (sum of the respective RA SGV#'s), the odds ratio (OR), the 95% confidence interval of the OR (lower limit "OR L95"; and upper limit "OR U95") and the probability (p-value "P regr") of the logistic regression analysis including the main effect of the single SGVs as covariate.

[0661] Table 11A. Combinations of 2 SGVs in subgroup of individuals comprising 2(+ / +) alleles of APOEe4 Odds ratio's (OR) of combinations of 2 SGVs selected from the APOEe4 + / + subgroup after stratification of the EADB dataset into 3 subgroups (APOEe4 APOEe4 + / -; APOEe4 + / +). SGV1 is the SGV1 id from the variant 1 in the combination, combining locus name and rsID; SGV2 is the SGV2 id from the variant 2 in the combination; RA SGV1 is the number of risk alleles in SGV1; RA SGV2 is the number of risk alleles in SGV2. Further included for each listed combination are the total RA (sum of the respective RA SGV#'s), the odds ratio (OR), the 95% confidence interval of the OR (lower limit "OR L95"; and upper limit "OR U95") and the probability (p-value "P regr") of the logistic regression analysis including the main effect of the single SGVs as covariate.

[0662] Table 12. SGVs in linkage disequilibrium (LD) with the SGVs listed in Table 1

[0663] The SGVs as listed in Table 1 ("AD SGVs") are occurring in the column "SGV2" and identified by their "rs ID #" as in Table 1 (cross-reference to Table 1). The SGVs in LD ("LD SGVs") with an AD SGV are given in the column "SGV1". The LD SGVs include the AD SGV of column "SGV2"; in some cases there is only one LD SGV, which then is the same as the AD SGV. The column SGV2 further includes the number of SGVl's (i.e. LD SGVs) linked to and including a given SGV2 AD SGV. Column r2: LD score (for LD SGVs = AD SGVs this score is always 1). The LD-SGV "11_47412669_A_AGGCTACCAGG" in LD with SGV2 rsl0437655 refers to chromosome 11 position 47412669, at which the nucleotide "A" is substituted by "AGGCTACCAGG" (genetic insertion); likewise the LD-SGV "12_113151026_A_AT" in LD with SGV2 rs6489896 refers to chromosome 12 position 113151026, at which the nucleotide "A" is substituted by "AT".

Claims

CLAIMS1. A method of diagnosing a subject as Alzheimer disease (AD) patient or as potential AD patient, or as subject being at risk of developing AD, the method including the steps of: determining the presence or absence of a pathogenic allele of the individual short genetic variants (SGVs) of a SGV combination in a DNA sample obtained from a subject, wherein the SGV combination is chosen from the SGV combinations as listed in Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table IDA, or Table 11A; diagnosing a subject as AD patient or as potential AD patient, or as subject being at risk of developing AD, when a pathogenic allele is present in each of the individual SGVs of the SGV combination.

2. The method according to claim 1 wherein the presence or absence of a pathogenic allele of the individual SGVs of more than one SGV combination is determined, and wherein the subject is diagnosed as AD patient or as potential AD patient, or as subject being at risk of developing AD, when a pathogenic allele is present in each of the individual SGVs of each of the more than one SGV combination.

3. A method of diagnosing a subject as Alzheimer disease (AD) patient or as potential AD patient, or as subject to be at risk of developing AD, the method including the steps of: determining the number of pathogenic alleles of the SGVs of a SGV combination in a DNA sample obtained from a subject, wherein the SGV combination is chosen from the SGV combinations as listed in Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table IDA, or Table 11A; diagnosing a subject as AD patient or as potential AD patient, or as subject being at risk of developing AD, when in each of the individual SGVs of the SGV combination the number of pathogenic alleles as listed for the SGV combination chosen from Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table IDA, or Table 11A is present.

4. The method according to claim 3 wherein the number of pathogenic alleles of the individual SGVs of more than one SGV combination is determined, and wherein the subject is diagnosed as AD patient or as potential AD patient, or as subject being at risk of developing AD, when in each of the individual SGVs of the more than one SGV combination the number of pathogenic alleles as listed for the more than one SGV combination chosen from Table 3B, Table 4B, Table 5B, Table 6B, Table 7B, Table 8B, Table 9A, Table IDA, or Table 11A is present.

5. The method according to any one of claims 1 to 4 wherein the pathogenic allele of an individual SGV is substituted for a SGV in linkage disequilibrium with the pathogenic allele of the individual SGV.

6. A method of screening for AD drugs / medicaments, the method comprising:- culturing cells obtained from an AD patient, wherein the AD patient is selected based on a method according to any one of claims 1 to 5, and wherein the cultured cells are displaying an AD-related phenotype;- contacting the cultured cells with a candidate AD drug / medicament;- selecting as AD drug / medicament a candidate AD drug / medicament that is treating the AD- related phenotype.

7. A method of selecting a subject for treatment with an AD drug / medicament, or for inclusion in a clinical trial of a candidate AD drug / medicament, wherein the subject is selected based on a method according to any one of claims 1 to 5.

8. A computer program having instructions which when executed cause a computing or data processing system or device to perform a method according to any one of claims 1 to 7, or to perform a step of a method according to any one of claims 1 to 7.

9. A computer program having instructions which when executed cause a computing or data processing system or device to perform a method for selecting SGV combinations for diagnosing a subject as Alzheimer disease (AD) patient or as potential AD patient, or as subject being at risk of developing AD, the method comprising the steps of: a first selection of SGV combinations by multifactor dimensionality reduction (MDR) analysis on a training set of subjects wherein the training set comprises AD patients and healthy subjects, and wherein the SGV combinations are selected starting from set of single SGVs that are associated with AD; logistic regression (LR) analysis of the first selected SGV combinations wherein the logistic regression incorporates as covariates the main effects of the single SGVs of the selected SGV combinations; a second selection of SGV combinations from the logistic regression analysis; validating the second selected SGV combinations in a test set of subjects wherein the test set comprises AD patients and healthy subjects; wherein the validated second selected SGV combinations are synergistic in separating an AD patient or subject at risk of developing AD from a healthy subject.

10. The computer program according to claim 9 to perform the method is further including a first selection of SGV combinations by MDR analysis on two training sets of subjects and including LR analysis on first selected SGV combinations common to the two training sets of subjects.

11. The computer program according to claim 9 or 10 to perform the method is further including selection of validated SGV combinations stratified according to the number of pathogenic APOEE4 alleles present in a subject.

12. The computer program according to any one of claims 9 to 11 to perform the method with substituting a single SGV associated with AD for a SGV in linkage disequilibrium with the single SGV associated with AD.

13. A computing or data processing system or device, machine readable medium, or diagnostic kit comprising a means for carrying out or performing a method according to any one of claims 1 to 12, or for carrying out or performing a step of a method according to any one of claims 1 to 12.

Citation Information

Patent Citations

  • Rapid detection of nucleic acid sequences in a sample by labeling the sample

    EP0235726A2

  • Method and kit for polynucleotide assay including primer-dependant DNA polymerase

    US4851331A

  • Methods and compositions for the detection of sequences in selected DNA molecules

    US5137806A

  • Method of detecting nucleotide sequences

    US5595890A

  • Allele specific polymerase chain reaction

    US5639611A