Biomarkers for diagnosing nonalcoholic steatohepatitis (NASH) or hepatocellular carcinoma (HCC)
Patent Information
- Application Number
- JP2023555119
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2021-09-30
- Filing Date
- 2021-11-24
- Publication Date
- 2024-12-03
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Current diagnostic techniques for non-alcoholic steatohepatitis (NASH) and hepatocellular carcinoma (HCC) lack the accuracy needed to precisely determine the stage of fatty liver disease (FLD) progression, failing to distinguish between different conditions such as fat deposits, NASH, and HCC, which hinders personalized treatment strategies.
A method combining site-specific glycoprotein analysis with machine learning and advanced mass spectrometry to quantify peptide structures, generating disease indicators for accurate classification of FLD progression stages, including NASH and HCC, using a supervised machine learning model to analyze peptide structural data from biological samples.
Enables precise diagnosis and tailored treatment plans by accurately distinguishing between healthy, NASH, HCC, and non-NASH/HCC conditions, allowing for timely intervention and improved patient outcomes.
Smart Images

Figure 2022115574000001 
Figure 2022115574000002 
Figure 2022115574000003
Abstract
Description
[Technical Field]
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to U.S. Provisional Patent Application No. 63 / 251,021, filed September 30, 2021, and U.S. Provisional Patent Application No. 63 / 118,486, filed November 25, 2020, each of which is incorporated by reference herein in its entirety.
[0002] The present disclosure relates generally to methods and systems for diagnosing and / or treating states of fatty liver disease (FLD) progression. More specifically, the present disclosure relates to analyzing quantification data of a set of peptide structures detected in a biological sample obtained from a subject for use in diagnostic assessment of the subject's disease state (e.g., healthy, NASH, HCC) associated with disease progression and / or for treating the subject. [Background technology]
[0003] Protein glycosylation and other post-translational modifications play an important role in virtually all aspects of human physiology. Not surprisingly, incomplete or altered protein glycosylation is often associated with various disease states. Identifying abnormal glycosylation provides opportunities for early detection, intervention, and treatment of affected subjects. Current biomarker identification methods, such as those developed in the fields of proteomics and genomics, can be used to detect indicators of certain diseases, such as cancer, and to distinguish certain types of cancer from other non-cancerous diseases. However, identifying disease processes using glycoproteomic analysis has not been successful to date. Furthermore, identifying disease states associated with disease progression using glycoproteomic analysis has not been successful to date.
[0004] Glycoprotein analysis is challenging on several levels. For example, a single glycan composition in a peptide may contain numerous isomeric structures resulting from different glycosidic linkages, branching patterns, and / or multiple monosaccharides with the same mass. In addition, the presence of multiple glycans sharing the same peptide backbone may result in assay signals from various glycoforms, whose individual abundances may be reduced compared to deglycosylated peptides. Therefore, the development of algorithms capable of identifying glycan structures on peptide fragments remains elusive.
[0005] In light of the above, there is a need for improved analytical methods involving site-specific analysis of glycoproteins to obtain information about protein glycosylation patterns, which in turn provides quantitative information that can be used to identify disease processes. The present disclosure addresses this and other needs by combining site-specific glycoprotein analysis with machine learning and advanced mass spectrometry instrumentation to quantitatively analyze peptide structures indicative of specific disease states, including, but not limited to, NASH and HCC.
[0006] Nonalcoholic fatty liver disease (NAFLD or FLD) is a by-product of the global epidemic of obesity and metabolic syndrome. FLD progresses through stages of fat accumulation and inflammation to NASH, and a small percentage of NASH patients progress to HCC. Knowing a patient's stage of FLD with high accuracy would allow healthcare professionals to customize treatment for each individual patient and achieve better outcomes. However, current diagnostic technologies lack the accuracy necessary to definitively predict the stage of FLD (e.g., whether a patient simply has fat accumulation, NASH, or HCC). Therefore, it may be desirable to have methods and systems that can distinguish these conditions from a healthy state. Summary of the Invention
[0007] In one or more embodiments, a method is provided for classifying a biological sample with respect to a plurality of conditions associated with fatty liver disease (FLD) progression. The method includes receiving peptide structure data corresponding to a set of glycoproteins in a biological sample obtained from a subject. Quantitation data identified from the peptide structure data for the set of peptide structures is input into a machine learning model. The set of peptide structures includes at least one peptide structure identified from the plurality of peptide structures in Table 1. The quantification data is analyzed using the machine learning model to generate a disease indicator. A diagnostic output is generated based on the disease indicator that classifies the biological sample as evidencing a corresponding condition among the plurality of conditions associated with FLD progression.
[0008] In one or more embodiments, a method is provided for training a model to diagnose a subject with one of a plurality of conditions associated with fatty liver disease (FLD) progression. The method includes receiving quantification data for a panel of peptide structures for a plurality of subjects diagnosed with a plurality of conditions associated with FLD progression. The quantification data includes a plurality of peptide structure profiles for the plurality of subjects, and identifies a corresponding state of the plurality of conditions for each peptide structure profile of the plurality of peptide structure profiles. A machine learning model is trained using the quantification data to determine which state of the plurality of conditions a biological sample from the subject corresponds to.
[0009] In one or more embodiments, a method is provided for detecting the presence of one of a plurality of conditions associated with fatty liver disease (FLD) progression in a biological sample. The method includes receiving peptide structure data corresponding to a set of glycoproteins in the biological sample obtained from a subject. The peptide structure data is analyzed using a supervised machine learning model to generate a disease indicator based on at least three peptide structures selected from the group of peptide structures identified in Table 1. In response to determining that the disease indicator falls within a selected range associated with the corresponding condition, the presence of the corresponding condition among the plurality of conditions associated with FLD progression is detected.
[0010] In one or more embodiments, a method for classifying a biological sample as corresponding to one of a plurality of states associated with fatty liver disease (FLD) progression is provided. The method includes training a supervised machine learning model using training data. The training data includes a plurality of peptide structure profiles of a plurality of training subjects, and identifies one state of a plurality of states for each of the plurality of peptide structure profiles. Peptide structure data corresponding to a set of glycoproteins in a biological sample obtained from the subject is received. Quantification data identified from the peptide structure data for the set of peptide structures is input into the trained supervised machine learning model. The set of peptide structures includes at least one peptide structure identified in Table 1. The quantification data is analyzed using the supervised machine learning model to generate a score. A determination is made that the score falls within a selected range associated with a corresponding state of the plurality of states associated with FLD progression. A diagnostic output is generated, the diagnostic output indicating that the biological sample demonstrates the corresponding state. The plurality of states include a non-alcoholic steatohepatitis (NASH) state, a hepatocellular carcinoma (HCC) state, and a non-NASH / HCC state.
[0011] In one or more embodiments, a method is provided for treating a nonalcoholic steatohepatitis (NASH) disorder in a patient to at least one of reduce, stall, or reverse the progression of the NASH disorder to hepatocellular carcinoma. The method includes receiving a biological sample from the patient. The amount of each peptide structure identified in Table 1 in the biological sample is determined using a multiple reaction monitoring mass spectrometry (MRM-MS) system. The amount of each peptide structure is analyzed using a machine learning model to generate a disease indicator. A diagnostic output is generated based on the disease indicator classifying the biological sample as evidence that the patient has a NASH disorder. Obeticholic acid (OCA) or a derivative thereof is administered to the patient. The administration includes at least one of intravenous and oral administration in the range of 10 to 25 mg daily.
[0012] In one or more embodiments, a method for treating hepatocellular carcinoma (HCC) disorder in a patient is provided. The method includes receiving a biological sample from the patient. The amount of each peptide structure identified in Table 1 in the biological sample is determined using a multiple reaction monitoring mass spectrometry (MRM-MS) system. A machine learning model is used to analyze the amount of each peptide structure to generate a disease indicator. A diagnostic output is generated based on the disease indicator, which classifies the biological sample as evidence that the patient has an HCC disorder. Treatment is administered, the treatment including administering to the patient sorafenib or a derivative thereof, wherein the administration includes at least one of intravenous and oral administration in a range of 775 to 825 mg daily; lenvatinib or a derivative thereof, wherein the administration includes at least one of intravenous and oral administration in a range of 7.5 to 8.5 mg / day when the patient weighs less than 60 kg, and at least one of intravenous and oral administration in a range of 11.5 to 12.5 mg / day when the patient weighs more than 60 kg; nivolumab or a derivative thereof, wherein the administration includes at least one of intravenous and oral administration in a range of 0.75 to 1.25 mg / day when the patient weighs more than 60 kg; g), regorafenib or a derivative thereof to the patient (which administration includes oral administration in the range of 150-170 mg / day), cabozantinib or a derivative thereof to the patient (which administration includes intravenous or oral administration in the range of 50-70 mg / day), or ramucirumab or a derivative thereof to the patient (which administration includes intravenous or oral administration in the range of 8-12 mg / kg).
[0013] In one or more embodiments, a method is provided for designing a treatment for a subject diagnosed with a disease state associated with fatty liver disease (FLD) progression, the method comprising, in response to determining that a biological sample obtained from the subject evidences the disease state using some or all of any one or more of the methods disclosed herein, designing a therapeutic agent for treating the subject.
[0014] In one or more embodiments, a method is provided for planning treatment of a subject diagnosed with a disease state associated with fatty liver disease (FLD) progression, the method comprising generating a treatment plan for treating the subject in response to determining that a biological sample obtained from the subject evidences the disease state using some or all of any one or more of the methods disclosed herein.
[0015] In one or more embodiments, a method is provided for treating a subject diagnosed with a disease state associated with fatty liver disease (FLD) progression, the method comprising administering to the subject a therapeutic agent to treat the subject based on determining that a biological sample obtained from the subject evidences the disease state using some or all of any one or more of the methods disclosed herein.
[0016] In one or more embodiments, a method is provided for treating a subject diagnosed with a disease condition associated with fatty liver disease (FLD) progression, the method comprising selecting a therapeutic agent for treating the subject based on determining that the subject will respond to the therapeutic agent using some or all of any one or more of the methods disclosed herein.
[0017] In one or more embodiments, a method for analyzing a set of peptide structures in a sample from a patient is provided. The method includes: (a) obtaining a sample from the patient; (b) preparing the sample to form a prepared sample including the set of peptide structures; (c) inputting the prepared sample into a monitoring mass spectrometry system using a liquid chromatography system; (d) detecting a set of product ions associated with each peptide structure of the set of peptide structures; and (e) generating quantification data for the set of product ions using the monitoring mass spectrometry system. The set of peptide structures includes at least one peptide structure selected from peptide structures PS-1 through PS-53 identified in Table 4. The set of peptide structures includes (i) a precursor ion having a mass-to-charge (m / z) ratio within ±1.5 of the m / z ratio listed for the precursor ion in Table 4 corresponding to the peptide structure; and (ii) a product ion having an m / z within ±1.0 of the m / z ratio listed for the first product ion in Table 4 corresponding to the peptide structure.
[0018] In one or more embodiments, a composition is provided, the composition comprising at least one of the peptide structures PS-1 through PS-53 identified in Table 1.
[0019] In one or more embodiments, a composition is provided, the composition comprising a peptide structure or product ion. The peptide structure or product ion comprises an amino acid sequence having at least 90% sequence identity to any one of SEQ ID NOS: 23-52, which correspond to peptide structures PS-1 through PS-53 in Table 1. The product ion is selected from the group consisting of the product ions identified in Table 4, including product ions that fall within the specified m / z ranges.
[0020] In one or more embodiments, a composition is provided, the composition comprising a glycopeptide structure selected as one from the group consisting of peptide structures PS-1 through PS-53 identified in Table 4. The glycopeptide structure comprises an amino acid peptide sequence identified in Table 5 as corresponding to the glycopeptide structure and a glycan structure identified in Table 1 as corresponding to the glycopeptide structure, wherein the glycan structure is linked to residues of the amino acid peptide sequence at corresponding positions identified in Table 1. The glycan structure has a glycan composition.
[0021] In one or more embodiments, a composition is provided, the composition comprising a peptide structure selected from a plurality of peptide structures identified in Table 1. The peptide structure has a monoisotopic mass identified as corresponding to the peptide structure in Table 1. The peptide structure comprises an amino acid sequence of SEQ ID NOs: 23-52 identified in Table 1 as corresponding to the peptide structure.
[0022] In one or more embodiments, a kit is provided that includes at least one agent for quantifying at least one peptide structure identified in Table 1 for performing part or all of any one or more of the methods disclosed herein.
[0023] In one or more embodiments, a kit is provided that includes at least one of a glycopeptide standard, a buffer, or a set of peptide sequences for performing any one or more of the methods disclosed herein, wherein a peptide sequence in the set of peptide sequences is identified by a corresponding one of SEQ ID NOS: 23-52 as defined in Table 1.
[0024] In one or more embodiments, a system includes one or more data processors and a non-transitory computer-readable storage medium containing instructions that, when executed on the one or more data processors, cause the one or more data processors to perform a portion or all of any one or more of the methods disclosed herein.
[0025] In one or more embodiments, a computer program product tangibly embodied in a non-transitory machine-readable storage medium is provided that includes instructions configured to cause one or more data processors to perform some or all of any one or more of the methods disclosed herein.
[0026] The present disclosure is described in conjunction with the accompanying drawings. [Brief explanation of the drawings]
[0027] [Figure 1] FIG. 1 is a schematic diagram of an exemplary workflow for detecting peptide structures associated with disease states for use in diagnosis and / or treatment, according to one or more embodiments. [Figure 2A] FIG. 1 is a schematic diagram of a preparation workflow according to one or more embodiments. [Figure 2B] FIG. 1 is a schematic diagram of data acquisition in accordance with one or more embodiments. [Figure 3] FIG. 1 is a block diagram of an analysis system according to one or more embodiments. [Figure 4] FIG. 1 is a block diagram of a computer system according to various embodiments. [Figure 5] 1 is a flowchart of a process for evaluating a biological sample obtained from a subject for FLD progression, according to one or more embodiments. [Figure 6] 1 is a flowchart of a process for detecting the presence of a disease state associated with FLD progression, according to one or more embodiments. [Figure 7] 1 is a flowchart of a process for training a supervised machine learning model to determine which of a plurality of states a biological sample corresponds to, according to various embodiments. [Figure 8] 1 is a flowchart of a process for classifying a biological sample as corresponding to one of a plurality of conditions associated with fatty liver disease (FLD) progression, according to one or more embodiments. [Figure 9]1 is a flowchart of a process for treating a subject for NASH, according to one or more embodiments. [Figure 10] 1 is a flowchart of a process for treating a subject for NASH, according to one or more embodiments. [Figure 11] 1 is a table of sample populations used in experiments, according to one or more embodiments. [Figure 12] 1 is a table of sample populations used in validation, according to one or more embodiments. [Figure 13] 1 is an illustration of a plot of the average normalized abundance of selected peptide structures identified and quantified via mass spectrometry as shown, according to one or more embodiments. [Figure 14] 1 is an illustration of a plot of normalized abundance of selected peptide structures of A2MG, according to one or more embodiments. [Figure 15] 1 is an illustration of a plot of normalized abundance of selected peptide structures of AGP1, according to one or more embodiments. [Figure 16] 1 is an illustration of a plot of normalized abundance of selected peptide structures of HPT, according to one or more embodiments. [Figure 17] 1 is an illustration of a plot of normalized abundance of selected peptide structures of CFAH, according to one or more embodiments. [Figure 18] 1 is an illustration of a plot of normalized abundance of selected peptide structures of A1AT, according to one or more embodiments. [Figure 19] 1 is a plot illustrating validation of the ability of a disease indicator to distinguish between control, NASH, and HCC states, according to one or more embodiments. [Figure 20] FIG. 1 is a plot of receiver operating characteristic (ROC) curves for distinguishing between NASH and HCC conditions for both the training and test sets, according to one or more embodiments. [Figure 21] 1 is a plot of the normalized abundance of these 10 peptide structures for control (eg, healthy) conditions versus HCC, and benign liver masses versus HCC, according to one or more embodiments. [Figure 22] 1 is a plot of receiver operating characteristic (ROC) curves for distinguishing between HCC and benign liver masses for both the training and test sets, according to one or more embodiments. DETAILED DESCRIPTION OF THE INVENTION
[0028] I. Overview The embodiments described herein recognize that glycoproteomics is an emerging field that can be used for the holistic diagnosis and / or treatment of subjects with various types of diseases. Glycoproteomics aims to determine the location, identity, and quantity of glycans and glycosylated proteins in a given sample (e.g., a blood sample, cells, tissue, etc.). Protein glycosylation is one of the most common and most complex forms of post-translational protein modification and can affect protein structure, conformation, and function. For example, glycoproteins can play important roles in important biological processes such as cell signaling, host-pathogen interactions, and immune responses and diseases. Therefore, glycoproteins can be important for diagnosing different types of diseases. Glycoproteins can also be important for distinguishing disease stages within diseases (e.g., FLD stages).
[0029] While protein glycosylation provides useful information for the staging of cancer and other diseases and disease analysis, protein glycosylation can be challenging because glycans typically cannot be traced back to their protein site of origin using currently available methodologies. Glycoprotein analysis can generally be challenging for several reasons. For example, a single glycan composition in a peptide can contain multiple isomeric structures due to different glycosidic linkages, branching, and many monosaccharides with the same mass. Furthermore, the presence of multiple glycans sharing the same peptide sequence may result in a split mass spectrometry (MS) signal into various glycoforms, whose individual abundances may be reduced compared to unglycosylated peptides (deglycosylated peptides).
[0030] However, to understand various disease states and disease progression and to more accurately diagnose certain disease states, it may be important to perform glycoprotein analysis and identify not only the glycans within a protein but also the attachment site (e.g., the amino acid residue at the attachment point). Therefore, there is a need to provide methods for site-specific glycoprotein analysis to obtain detailed information about protein glycosylation patterns, which may provide information about the disease state. This information can be used to distinguish a disease state from other conditions, diagnose whether a subject has the disease state, determine the likelihood that a subject has the disease state, or a combination thereof. Such analysis may be useful, for example, but not limited to, for distinguishing between two or more of a non-alcoholic steatohepatitis (NASH) condition, a hepatocellular carcinoma (HCC) condition, and a non-NASH / HCC condition (which may include at least one of a non-alcoholic FLD disease state, a control condition, a healthy condition, a condition without liver disease, or a benign liver mass condition).
[0031] Accordingly, embodiments described herein provide various methods and systems for analyzing proteins, particularly glycoproteins, in a subject. In one or more embodiments, a machine learning model is trained to analyze peptide structure data and generate disease indicators that provide information about one or more diseases. For example, in various embodiments, the peptide structure data includes quantification metrics (e.g., abundance or concentration data) for the peptide structures. A peptide structure may be defined by a deglycosylated peptide sequence (e.g., a peptide or peptide fragment of a larger parent protein) or a glycosylated peptide sequence. A glycosylated peptide sequence (also referred to as a glycopeptide structure) may be a peptide sequence having a glycan structure attached to a linkage site (e.g., an amino acid residue) of the peptide sequence, which may occur, for example, through a specific atom of the amino acid residue. Non-limiting examples of glycosylated peptides include N-linked glycopeptides and O-linked glycopeptides.
[0032] The embodiments described herein recognize that the abundance of selected peptide structures in a biological sample obtained from a subject may be used to determine the likelihood that the subject has a particular disease state (e.g., stage of FLD).
[0033] Analyzing the abundance of peptide sequences and glycosylated peptide sequences in biological samples can provide a more accurate method for distinguishing between progression states within FLD. This type of peptide structure analysis can help generate an accurate diagnosis compared to glycoprotein analysis, which focuses on analyzing glycoproteins that are too large to be resolved by mass spectrometry. Furthermore, glycoproteins may have too many potential proteoforms to consider. Furthermore, analyzing peptide structure data in the manner described by various embodiments herein can help generate an accurate diagnosis compared to glycomics analysis, which provides little or no information about which proteins and which amino acid residues various glycan structures are attached to.
[0034] The following description provides exemplary implementations of the methods and systems described herein for the study, diagnosis, and / or treatment (e.g., treatment design, planning, and / or manufacture) of disease states associated with FLD (e.g., NASH states, HCC states, etc.). Explanations and examples of various terms as used herein are provided in Section II below.
[0035] II. Illustrative Explanation of Terms The term "ones" means more than one.
[0036] As used herein, the term "plurality" may be 2, 3, 4, 5, 6, 7, 8, 9, 10, or more.
[0037] As used herein, the term "set" means one or more. For example, a set of items includes one or more items.
[0038] As used herein, the phrase “at least one of,” when used in conjunction with a list of items, means that different combinations of one or more of the listed items may be used, and that only one of the items in the list may be required. An item may be a specific object, thing, step, operation, process, or category. In other words, “at least one of” means that any combination or number of items from the list may be used, but not all of the items in the list may be required. For example, without limitation, “at least one of item A, item B, or item C” means item A; item A and item B; item B; item A, item B, and item C; item B and item C; or item A and C. In some cases, “at least one of item A, item B, or item C” means, but is not limited to, two of item A, one of item B, and ten of item C; four of item B and seven of item C; or some other suitable combination.
[0039] As used herein, "substantially" means sufficient to operate for its intended purpose. Thus, the term "substantially" allows for slight, inconsequential variations from absolute or perfect conditions, dimensions, measurements, results, etc., as would be expected by one of ordinary skill in the art, which do not appreciably affect overall performance. When used in reference to a number or parameter or characteristic that can be expressed as a number, "substantially" means within 10%.
[0040] The term "amino acid," as used herein, generally refers to any organic compound containing an amino group (e.g., -NH), a carboxyl group (-COOH), and a side group (R), which varies based on the particular amino acid. Amino acids may be linked using peptide bonds.
[0041] The term "alkylation," as used herein, generally refers to the transfer of an alkyl group from one molecule to another. In various embodiments, alkylation is used to react with reduced cysteines to prevent the reformation of disulfide bonds after reduction has occurred.
[0042] The term "linkage site" or "glycosylation site," as used herein, generally refers to a location where a sugar molecule of a glycan or glycan structure is directly attached (e.g., covalently attached) to an amino acid of a peptide, polypeptide, or protein. For example, the linkage site can be an amino acid residue, and the glycan structure can be attached via an atom of the amino acid residue. Non-limiting examples of types of glycosylation include N-linked glycosylation, O-linked glycosylation, C-linked glycosylation, S-linked glycosylation, and glycation.
[0043] The terms "biological sample," "biological specimen," or "biological specimen," as used herein, generally refer to a specimen obtained by sampling, typically from a subject, to represent the source of the specimen. A biological sample may be representative of a whole organism, a specific tissue, a cell type, or a category or subcategory of interest. A biological sample may include macromolecules. A biological sample may include small molecules. A biological sample may include viruses. A biological sample may include cells or derivatives of cells. A biological sample may include organelles. A biological sample may include cell nuclei. A biological sample may include rare cells from a population of cells. A biological sample may include any type of cell, including, but not limited to, prokaryotic cells, eukaryotic cells, bacteria, fungi, plants, mammalian, or other animal cell types, mycoplasma, healthy tissue cells, tumor cells, or any other cell type, whether from a unicellular or multicellular organism. A biological sample may include cellular components. A biological sample may include nucleotides (e.g., ssDNA, dsDNA, RNA), organelles, amino acids, peptides, proteins, carbohydrates, glycoproteins, or any combination thereof. A biological sample may include cells, or a matrix (e.g., a gel or polymer matrix) containing one or more components from cells (e.g., cell beads), such as DNA, RNA, organelles, proteins, or any combination thereof from cells. A biological sample may be obtained from a subject's tissue. A biological sample may include cured cells. Such cured cells may or may not include cell walls or cell membranes. A biological sample may include one or more components of cells, but may not include other components of cells. Examples of such components may include nuclei or organelles. A biological sample may not include living cells. Living cells may be culturable.
[0044] The term "biomarker," as used herein, generally refers to any measurable substance sampled from a subject, the presence of which indicates some phenomenon. Non-limiting examples of such phenomena can include a disease state, a medical condition, or exposure to a compound or environmental condition. In various embodiments described herein, biomarkers may be used for diagnostic purposes (e.g., to diagnose a disease state, a healthy state, an asymptomatic state, a symptomatic state, etc.). The term "biomarker" may be used interchangeably with the term "marker."
[0045] The term "denatured," as used herein, generally refers to any molecule that loses the quaternary, tertiary, and secondary structure present in its native state. Non-limiting examples include proteins or nucleic acids that have been exposed to external compounds or environmental conditions, such as acid, base, temperature, pressure, radiation, etc.
[0046] The term "denatured protein," as used herein, generally refers to a protein that loses the quaternary, tertiary, and secondary structure present in its native state.
[0047] The terms "digestion" or "enzymatic digestion," as used herein, generally refer to breaking down a polymer (e.g., cleaving a polypeptide at a cleavage site). Proteins may be digested in preparation for mass spectrometry using a trypsin digestion protocol. If access is limited to the cleavage site, proteins may also be digested using other proteases in preparation for mass spectrometry.
[0048] The terms "immune checkpoint inhibitor therapeutic" and "immune checkpoint inhibitor drug," as used herein, generally refer to drugs or therapeutics that can target immune checkpoint molecules (e.g., molecules on immune cells that need to be activated (or inactivated) to initiate an immune response). Non-limiting examples of immune checkpoint inhibitor therapeutics can include pembrolizumab, nivolumab, and cemiplimab.
[0049] The term "disease progression," as used herein, refers to the progression of a disease from no disease or a less advanced (e.g., severe) form of the disease to a more advanced (e.g., severe) form of the disease. Disease progression may include any number of stages of the disease.
[0050] The term "disease state," as used herein, generally refers to a condition that affects the structure or function of an organism. Non-limiting examples of causes of a disease state can include pathogens, immune system dysfunction, cell damage caused by aging, and cell damage caused by other factors (e.g., trauma and cancer). A disease state can include, for example, stages of disease progression. For example, in the case of FLD, progression can progress from a healthy stage to fat accumulation and inflammation (fatty liver), nonalcoholic steatohepatitis (NASH), fibrosis, and cirrhosis. In some cases, progression can progress from NASH to hepatocellular carcinoma (HCC). A disease state can include any stage of disease, whether symptomatic or asymptomatic. A disease state can cause mild, moderate, or severe disruption to the structure or function of a subject.
[0051] The terms "glycan" and "polysaccharide," as used herein, both generally refer to the carbohydrate residue of a glycoconjugate, e.g., the carbohydrate moiety of a glycopeptide, glycoprotein, glycolipid, or proteoglycan. Glycans can include monosaccharides.
[0052] The term "glycopeptide" or "glycopolypeptide," as used herein, generally refers to a peptide or polypeptide that includes at least one glycan residue. In various embodiments, a glycopeptide includes a carbohydrate moiety (e.g., one or more glycans) covalently attached to the side chain (i.e., R group) of an amino acid residue.
[0053] The term "glycoprotein," as used herein, generally refers to a protein having at least one glycan residue attached thereto. In some instances, a glycoprotein is a protein having at least one covalently attached oligosaccharide chain. Examples of glycoproteins include apolipoprotein C-III (APOC3), alpha-1-antichymotrypsin (AACT), afamin (AFAM), alpha-1-acid glycoprotein 1 and 2 (AGP12), apolipoprotein B-100 (APOB), apolipoprotein D (APOD), complement C1s subcomponent (C1S), calpain-3 (CAN3), clusterin (CLUS), complement component C8A chain (CO8A), alpha-2-HS glycoprotein, and the like. Examples of glycopeptides include, but are not limited to, protein (FETUA), haptoglobin (HPT), immunoglobulin heavy constant gamma 1 (IgG1), immunoglobulin J chain (IgJ), plasma kallikrein (KLKB1), serum paraoxonase / arylesterase 1 (PON1), prothrombin (THRB), serotransferrin (TRFE), protein unc-13 homolog A (UN13A), and zinc-alpha-2-glycoprotein (ZA2G). Glycopeptide, as used herein, unless otherwise specified, refers to a fragment of a glycoprotein.
[0054] The term "liquid chromatography," as used herein, generally refers to a technique used to subdivide a sample. Liquid chromatography can be used to separate, identify, and quantify components.
[0055] The term "mass spectrometry," as used herein, generally refers to an analytical technique used to identify molecules. In various embodiments described herein, mass spectrometry may be involved in the characterization and sequencing of proteins.
[0056] The term "m / z" or "mass-to-charge ratio," as used herein, generally refers to the output value from a mass spectrometry instrument. In various embodiments, m / z can represent the relationship between the mass of a given ion and the number of elementary charges it carries. The "m" in m / z represents mass and the "z" represents charge. In some embodiments, m / z can be displayed on the x-axis of a mass spectrum.
[0057] The term "peptide," as used herein, generally refers to amino acids linked by peptide bonds. A peptide can include an amino acid chain of 10 to 50 residues. A peptide can include an amino acid chain shorter than 10 residues, including oligopeptides, dipeptides, tripeptides, and tetrapeptides. A peptide can include a chain longer than 50 residues and may be referred to as a "polypeptide" or a "protein."
[0058] The terms "protein" or "polypeptide" or "peptide" may be used interchangeably herein and generally refer to molecules containing at least three amino acid residues. Proteins may include polymer chains made of amino acid sequences linked together by peptide bonds. Proteins may be digested for mass spectrometry using a trypsin digestion protocol. Proteins may also be digested using other proteases for mass spectrometry if access is limited to the cleavage site.
[0059] The term "peptide structure," as used herein, generally refers to a peptide or a portion thereof, or a glycopeptide or a portion thereof. In various embodiments described herein, a peptide structure can include any molecule that includes at least two amino acids in sequence.
[0060] The term "reduction," as used herein, generally refers to the gain of electrons by a substance. In various embodiments described herein, a sugar can be directly attached to a protein, thereby reducing the amino acid to which it is attached. Such a reduction reaction can occur in glycosylation. In various embodiments, reduction can be used to break a disulfide bond between two cysteines.
[0061] The term "sample," as used herein, generally refers to a sample from a subject of interest and may include a biological sample of the subject. A sample may include a cell sample. A sample may include a cell line or cell culture sample. A sample may include one or more cells. A sample may include one or more microorganisms. A sample may include a nucleic acid sample or a protein sample. A sample may also include a carbohydrate sample or a lipid sample. A sample may be derived from another sample. A sample may include a tissue sample, such as a biopsy, core biopsy, needle aspirate, or fine needle aspirate. A sample may include a fluid sample, e.g., a blood sample, a urine sample, or a saliva sample. A sample may include a skin sample. A sample may include a buccal swab. A sample may include a plasma or serum sample. A sample may include a cell-free or acellular sample. An acellular sample may include extracellular polynucleotides. The sample may be derived from blood, plasma, serum, urine, saliva, mucosal discharge, sputum, stool, or tears. The sample may be derived from red blood cells or white blood cells. The sample may be derived from feces, spinal fluid, CNS fluid, gastric fluid, amniotic fluid, cyst fluid, peritoneal fluid, bone marrow, bile, other body fluids, tissue obtained from a biopsy, skin, or hair.
[0062] The term "sequence," as used herein, generally refers to a biological sequence comprising one-dimensional monomers that can be assembled to form a polymer. Non-limiting examples of sequences include nucleotide sequences (e.g., ssDNA, dsDNA, and RNA), amino acid sequences (e.g., proteins, peptides, and polypeptides), and carbohydrates (e.g., Cm (H2O) n (Compounds containing
[0063] The term "subject," as used herein, generally refers to an animal, such as a mammal (e.g., a human) or an avian (e.g., a bird), or other organism, such as a plant. For example, a subject may include a vertebrate, a mammal, a rodent (e.g., a mouse), a primate, a monkey, or a human. Animals may include, but are not limited to, farm animals, sport animals, and pets. A subject may include a healthy or asymptomatic individual, an individual suspected or suspected of having a disease (e.g., cancer) or a predisposition to a disease, and / or an individual in need of therapy or suspected of being in need of therapy. A subject may be a patient. A subject may include a microorganism or fungus (e.g., bacteria, fungi, archaea, virus).
[0064] As used herein, a "model" may include one or more algorithms, one or more mathematical techniques, one or more machine learning algorithms, or a combination thereof.
[0065] As used herein, "machine learning" may refer to the act of using algorithms to analyze data, learn from it, and then make decisions or predictions about things in the world. Machine learning uses algorithms that can learn from data without relying on rule-based programming. Machine learning algorithms may include parametric models, non-parametric models, deep learning models, neural networks, linear discriminant analysis models, quadratic discriminant analysis models, support vector machines, random forest algorithms, nearest neighbor algorithms, composite discriminant analysis models, k-means clustering algorithms, supervised models, unsupervised models, logistic regression models, multivariate regression models, penalized multivariate regression models, or another type of model.
[0066] As used herein, "artificial neural network" or "neural network" (NN) may refer to a mathematical algorithm or computational model that mimics an interconnected group of artificial nodes or neurons that process information based on a connectionist approach to computation. A neural network, also referred to as a neural net, can predict an output for a received input using one or more layers of nonlinear units. Some neural networks include one or more hidden layers in addition to an output layer. The output of each hidden layer is used as an input to the next layer of the network, i.e., the next hidden layer or output layer. Each layer of the network generates an output from the received input according to the current values of a respective set of parameters. In various embodiments, a reference to a "neural network" may refer to one or more neural networks.
[0067] A neural network can process information in two ways: when it is being trained, it is in training mode, and when it puts what it has learned into practice, it is in inference (or prediction) mode. Neural networks learn through a feedback process (e.g., backpropagation), which allows the network to adjust the weight coefficients of individual nodes in intermediate hidden layers (modify its behavior) so that its output matches the output of the training data. In other words, a neural network learns by being given training data (training examples) and eventually learns how to arrive at the correct output even when presented with a new range or set of inputs. The neural network may include, for example, without limitation, at least one of a feedforward neural network (FNN), a recurrent neural network (RNN), a modular neural network (MNN), a convolutional neural network (CNN), a residual neural network (ResNet), an ordinary differential equation neural network (neural-ODE), or another type of neural network.
[0068] As used herein, a "target glycopeptide analyte" may refer to a peptide structure (e.g., glycosylated or deglycosylated / unglycosylated), a portion of a peptide structure, a substructure of a peptide structure (e.g., a glycan or glycosylation site), a product of one or more of the structures and substructures listed above, an associated detection molecule (e.g., a signal molecule, label, or tag), or an amino acid sequence that can be measured by mass spectrometry.
[0069] As used herein, "peptide dataset" may be used interchangeably with "peptide structural data" and may refer to any data of or related to peptides from a resulting mass spectrometry run. A peptide dataset may include data obtained from a sample or biological sample using mass spectrometry. A peptide dataset may include data related to an NGEP external standard, data related to an internal standard, and data related to the target glycopeptide analyte of the sample. A peptide dataset may be obtained from analyses resulting from a single run. In some embodiments, a peptide dataset may include the raw abundance and mass-to-charge ratio of one or more peptides.
[0070] As used herein, "non-glycosylated endogenous peptide" ("NGEP") may also be referred to as a deglycosylated peptide and may refer to a peptide structure that does not contain glycan molecules. In various embodiments, the NGEP and the target glycopeptide analyte may originate from the same subject. In various embodiments, the NGEP may be isotopically labeled in preparation for mass spectrometry analysis.
[0071] As used herein, a "transition" may refer to or identify a peptide structure. In some embodiments, a transition may refer to a particular pair of m / z values associated with a precursor ion and a product or fragment ion.
[0072] As used herein, "non-glycosylated endogenous peptide" ("NGEP") may refer to a peptide structure that does not contain glycan molecules. In various embodiments, the NGEP and the target glycopeptide analyte may be derived from the same protein sequence. In some embodiments, the NGEP and the target glycopeptide analyte may be derived from or comprise the same peptide sequence. In various embodiments, the NGEP may be isotopically labeled in preparation for mass spectrometry.
[0073] As used herein, "abundance value" may refer to "abundance" or a quantitative value associated with abundance.
[0074] As used herein, "abundance" may refer to a quantitative value generated using mass spectrometry. In various embodiments, the quantitative value may relate to the amount of a particular peptide structure (e.g., a biomarker) present in a biological sample. In some embodiments, the amount may be relative to other structures present in the sample (e.g., relative abundance). In some embodiments, the quantitative value may include the amount of an ion generated using mass spectrometry. In some embodiments, the quantitative value may be associated with an m / z value (e.g., abundance on the x-axis and m / z on the y-axis). In other embodiments, the quantitative value may be expressed in atomic mass units.
[0075] As used herein, "relative abundance" may refer to a comparison of two or more abundances. In various embodiments, the comparison may include comparing one peptide structure to the total number of peptide structures. In some embodiments, the comparison may include comparing one peptide glycotype (e.g., two identical peptides that differ by one or more glycans) to a set of peptide glycotypes. In some embodiments, the comparison may include comparing the number of ions having a particular m / z ratio with the total number of ions detected. In various embodiments, the relative abundance may be expressed as a ratio. In other embodiments, the relative abundance may be expressed as a percentage. The relative abundance may be presented on the y-axis of a mass spectrum plot.
[0076] As used herein, "internal standard" may refer to something that can be contained (e.g., spiked in) in the same sample as the target glycopeptide analyte undergoing mass spectrometry. Internal standards can be used for calibration purposes. In addition, internal standards can be used in the systems and methods described herein. In some embodiments, internal standards can be selected based on similar m / z and / or retention time and can be "surrogates" if a particular standard is too expensive or unavailable. Internal standards can be heavily labeled or non-heavily labeled.
[0077] III. Example Workflow Overview 1 is a schematic diagram of an exemplary workflow 100 for detecting peptide structures associated with a disease state for use in diagnosis and / or treatment, according to one or more embodiments. Workflow 100 may include various operations, including, for example, sample collection 102, sample intake 104, sample preparation and processing 106, data analysis 108, and output generation 110.
[0078] Sample collection 102 may include obtaining a biological sample 112 from one or more subjects, such as, for example, subject 114. The biological sample 112 may take the form of a specimen obtained via one or more sampling methods. The biological sample 112 may be representative of the entire subject 114, or a particular tissue, cell type, or other category or subcategory of interest. The biological sample 112 may be obtained in any of a number of different ways. In various embodiments, the biological sample 112 comprises a whole blood sample 116 obtained via a blood draw. In other embodiments, the biological sample 112 comprises a set of aliquots 118 including, for example, a serum sample, a plasma sample, a blood cell (e.g., white blood cell (WBC) or red blood cell (RBC) sample), another type of sample, or a combination thereof. The biological sample 112 may include nucleotides (e.g., ssDNA, dsDNA, RNA), organelles, amino acids, peptides, proteins, carbohydrates, glycoproteins, or any combination thereof.
[0079] In various embodiments, a single run can analyze a sample (e.g., a sample containing a peptide analyte), an external standard (e.g., NGEP of a serum sample), and an internal standard. Thus, abundance values (e.g., abundance or raw abundance) of the external standard, internal standard, and target glycopeptide analyte can be determined by mass spectrometry in the same run.
[0080] In various embodiments, the external standard may be analyzed before analyzing the samples. In various embodiments, the external standard may be run independently between samples. In some embodiments, the external standard may be analyzed after every 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or more experiments. In various embodiments, the external standard data may be used in some or all of the normalization systems and methods described herein. In additional embodiments, blank samples may be processed to prevent column fouling.
[0081] Sample intake 104 may include one or more of a variety of operations, such as, for example, aliquoting, registering, processing, storing, thawing, and / or other types of operations. In one or more embodiments, when the biological sample 112 includes a whole blood sample 116, sample intake 104 may include aliquoting the whole blood sample 116 to form a set of aliquots, which may then be further aliquoted to form a set of samples 120.
[0082] Sample preparation and processing 106 may include, for example, one or more operations to form a set of peptide structures 122. In various embodiments, the set of peptide structures 122 may include various fragments of unfolded proteins that may have undergone digestion and are ready for analysis.
[0083] Additionally, sample preparation and processing 106 may include, for example, data mining 124 based on the set of peptide structures 122. For example, data mining 124 may include, for example, without limitation, the use of a liquid chromatography / mass spectrometry (LC / MS) system.
[0084] Data analysis 108 may include, for example, peptide structure analysis 126. In some embodiments, data analysis 108 also includes output generation 110. In other embodiments, output generation 110 may be considered a separate operation from data analysis 108. Output generation 110 may include, for example, generating final output 128 based on the results of peptide structure analysis 126. In various embodiments, final output 128 may be used to research, diagnose, and / or determine treatment for conditions associated with fatty liver disease.
[0085] In various embodiments, the final output 128 is comprised of one or more outputs. The final output 128 can take various forms. For example, the final output 128 may be a report including, for example, a diagnostic output, a treatment output (e.g., a treatment design output, a treatment plan output, or a combination thereof), analyzed data (e.g., relativized and normalized), or a combination thereof. In some embodiments, the report may include the target glycopeptide analyte concentration as a function of the NGEP concentration value and the normalized abundance value. In some embodiments, the final output 128 may be an alert (e.g., a visual alert, an audible alert, etc.), a notification (e.g., a visual notification, an audible notification, an email notification, etc.), an email output, or a combination thereof. In some embodiments, the final output 128 may be transmitted to a remote system 130 for processing. The remote system 130 may include, for example, a computer system, a server, a processor, a cloud computing platform, cloud storage, a laptop, a tablet, a smartphone, some other type of mobile computing device, or a combination thereof.
[0086] In other embodiments, workflow 100 may optionally exclude one or more of the operations described herein and / or may optionally include one or more other steps or operations other than those described herein (e.g., in addition to and / or instead of those described herein). Thus, workflow 100 may be implemented in any of a number of different ways for use in, for example, researching, diagnosing, and / or treating FLD.
[0087] IV. Detection and Quantification of Peptide Structure Figures 2A and 2B are schematic diagrams of a workflow for sample preparation and processing 106, according to one or more embodiments. Figures 2A and 2B are described with continued reference to Figure 1. Sample preparation and processing 106 may include, for example, a preparation workflow 200 shown in Figure 2A and a data retrieval 124 shown in Figure 2B.
[0088] IV.A. Sample Preparation and Processing 2A is a schematic diagram of a preparation workflow 200 according to one or more embodiments. The preparation workflow 200 may be used to prepare samples, such as samples in the set of samples 120 of FIG. 1, for analysis via data mining 124. For example, this analysis may be performed via mass spectrometry (e.g., LC-MS). In various embodiments, the preparation workflow 200 may include denaturation and reduction 202, alkylation 204, and digestion 206.
[0089] Generally, polymers, such as proteins, in their native form can fold to include secondary, tertiary, and / or other higher-order structures. Such higher-order structures may allow the protein to function and complete a task in a target (e.g., enable enzymatic activity). Furthermore, such higher-order structures of polymers may be maintained through various interactions between the side chains of amino acids within the polymer. Such interactions may include ionic bonds, hydrophobic interactions, hydrogen bonds, and disulfide bonds between cysteine residues. However, when using analytical systems and methods including mass spectrometry, it may be desirable to unfold such polymers (e.g., peptide / protein molecules) to obtain sequence information. In some embodiments, unfolding the polymer may include denaturing the polymer, which may include, for example, linearizing the polymer.
[0090] In one or more embodiments, denaturation and reduction 202 can be used to disrupt the higher order structure (e.g., secondary, tertiary, quaternary, etc.) of one or more proteins (e.g., polypeptides and peptides) in a sample (e.g., one of the sets of samples 120 of FIG. 1 ). Denaturation and reduction 202 can include, for example, a denaturation procedure and a reduction procedure. In some embodiments, the denaturation procedure can be performed using, for example, thermal denaturation, which uses heat as a denaturing agent. Thermal denaturation can disrupt ionic bonds, hydrophobic interactions, and / or hydrogen bonds.
[0091] In one or more embodiments, the denaturing procedure may include the use of one or more denaturing agents, temperature (e.g., heat), or both. These one or more denaturing agents may include, for example, but are not limited to, any number of chaotropic salts (e.g., urea, guanidine), detergents (e.g., sodium dodecyl sulfate (SDS), beta-octyl glucoside, Triton X-100), or combinations thereof. In some cases, such denaturing agents may be used in conjunction with heat when a decontamination procedure is further included in the sample preparation workflow.
[0092] The resulting one or more denatured (e.g., unfolded, linearized) proteins may then undergo further processing in preparation for analysis. For example, a reduction procedure may be performed in which one or more reducing agents are applied. In various embodiments, the reducing agent may produce an alkaline pH. The reducing agent may take the form of, for example, but not limited to, dithiothreitol (DTT), tris(2-carboxyethyl)phosphine (TCEP), or some other reducing agent. The reducing agent may reduce (e.g., cleave) disulfide bonds between cysteine residues of the one or more denatured proteins to form one or more reduced proteins.
[0093] In various embodiments, one or more reduced proteins resulting from denaturation and reduction 202 may be subjected to a process that prevents, for example, the reformation of disulfide bonds between cysteine residues of the one or more reduced proteins. This process may be carried out using alkylation 204 to form one or more alkylated proteins. For example, alkylation 204 may be used to add an acetamide group to the sulfur on each cysteine residue, preventing the reformation of disulfide bonds. In various embodiments, the acetamide groups may be added by reacting one or more alkylating agents with the reduced protein. The one or more alkylating agents may include, for example, one or more acetamide salts. The alkylating agent may take the form of, for example, iodoacetamide (IAA), 2-chloroacetamide, some other type of acetamide salt, or some other type of alkylating agent.
[0094] In some embodiments, alkylation 204 may include a quenching procedure, which may be performed using one or more reducing agents (e.g., one or more of the reducing agents described above).
[0095] In various embodiments, one or more alkylated proteins formed via alkylation 204 may then undergo digestion 206 in preparation for analysis (e.g., mass spectrometry). Protein digestion 206 may include cleaving the protein at or around one or more cleavage sites (e.g., site 205, which may be one or more amino acid residues). For example, without limitation, the alkylated protein may be cleaved at the carboxyl side of a lysine or arginine residue. This type of cleavage may divide the protein into various segments containing one or more peptide structures (e.g., glycosylated or deglycosylated).
[0096] In various embodiments, digestion 206 is performed using one or more proteolytic catalysts. For example, an enzyme can be used in digestion 206. In some embodiments, the enzyme takes the form of trypsin. In other embodiments, one or more other types of enzymes (e.g., proteases) can be used in addition to or instead of trypsin. These one or more other enzymes can include, but are not limited to, LysC, LysN, AspN, GluC, and ArgC. In some embodiments, digestion 206 can be performed using tosylphenylalanyl chloromethyl ketone (TPCK)-treated trypsin, one or more engineered forms of trypsin, one or more other formulations of trypsin, or combinations thereof. In some embodiments, digestion 206 can be performed in multiple steps, each involving the use of one or more digestion agents. For example, secondary digestion, tertiary digestion, etc. can be performed. In one or more embodiments, trypsin is used to digest serum samples. In one or more embodiments, a trypsin / LysC cocktail is used to digest plasma samples.
[0097] In some embodiments, digestion 206 further includes a quenching step, which may be performed by acidifying the sample (e.g., to a pH below 3). In some embodiments, this acidification may be performed using formic acid.
[0098] In various embodiments, preparation workflow 200 further includes a post-digestion procedure 207. Post-digestion procedure 207 may include, for example, a decontamination procedure. Decontamination procedures may include, for example, removal of unwanted components in the sample resulting from digestion 206. For example, unwanted components may include, but are not limited to, inorganic ions, surfactants, etc. In some embodiments, post-digestion procedure 207 further includes a procedure for adding a heavy-labeled peptide internal standard.
[0099] Although the preparation workflow 200 is described with respect to samples made or collected from blood-based biological samples 112 (e.g., whole blood samples, plasma samples, serum samples, etc.), the sample preparation workflow 200 may similarly be performed with other types of samples (e.g., tears, urine, tissue, interstitial fluid, sputum, etc.) to produce a set of peptide structures 122.
[0100] IV.B. Peptide Structure Identification and Quantification 2B is a schematic diagram of data acquisition 124 according to one or more embodiments. In various embodiments, data acquisition 124 may begin after sample preparation 200 described in FIG. 2A. In various embodiments, data acquisition 124 may include quantification 208, quality control 210, and peak integration and normalization 212.
[0101] In various embodiments, targeted quantification 208 of peptides and glycopeptides may incorporate the use of liquid chromatography-mass spectrometry (LC / MS) instrumentation. For example, LC-MS / MS, or tandem MS, may be used. Generally, LC / MS (e.g., LC-MS / MS) can combine the physical separation capabilities of liquid chromatography (LC) with the mass analysis capabilities of mass spectrometry (MS). According to some embodiments described herein, this technique allows for the separation of digested peptides fed from an LC column through an interface into an MS ion source.
[0102] In various embodiments, any LC / MS device can be incorporated into the workflow described herein. In various embodiments, an instrument or instrument system suitable for identification and targeted quantification 208 may include, for example, a Triple Quadrupole LC / MS™. In various embodiments, targeted quantification 208 is performed using multiple reaction monitoring mass spectrometry (MRM-MS).
[0103] In various embodiments described herein, the identity and associated abundance of a particular protein or peptide can be assessed. In various embodiments described herein, the identity and associated abundance of a particular glycan can be assessed. In various embodiments described herein, a particular glycan can be matched to a glycosylation site on a protein or peptide and an abundance value determined.
[0104] In some cases, targeted quantification 208 involves consistently identifying abundant product ions using specific collision energies associated with appropriate fragmentation. Glycopeptide structures may have lower collision energies than deglycosylated peptide structures. When analyzing samples containing glycopeptide structures, source voltages and gas temperatures may be reduced compared to typical proteomics analyses.
[0105] In various embodiments, quality control 210 procedures can be implemented to optimize data quality. In various embodiments, measurements can be implemented that only allow for errors within an acceptable range outside of expected values. In various embodiments, quality control 210 can be assisted by the use of statistical models (e.g., using Westgard rules). For example, quality control 210 can include, for example, assessing the retention times and abundances of representative peptide structures (e.g., glycosylated and / or deglycosylated) and spiked-in internal standards in either all samples or each quality control sample (e.g., pooled serum digests).
[0106] Peak integration and normalization 212 may be performed to process the generated data and convert the data into a format for analysis. For example, peak integration and normalization 212 may include converting the abundance data of the various product ions detected for a selected peptide structure into a single quantification metric for that peptide structure (e.g., relative amount, adjusted amount, normalized amount, relative concentration, adjusted concentration, normalized concentration, etc.). In some embodiments, peak integration and normalization 212 may be performed using one or more of the techniques described in U.S. Patent Publication No. 2020 / 0372973 A1 and / or U.S. Patent Publication No. 2020 / 0240996 A1, the disclosures of which are incorporated herein by reference in their entireties.
[0107] V. Exemplary Systems for Peptide Structure Data Analysis VA Analysis System for Peptide Structure Data Analysis Figure 3 is a block diagram of an analysis system 300 according to one or more embodiments. Analysis system 300 can be used to detect and analyze various peptide structures associated with different states of FLD. Analysis system 300 is one example of an implementation of a system that may be used to perform data analysis 108 of Figure 1. For this reason, analysis system 300 will be described with continued reference to workflow 100 described in Figures 1, 2A, and / or 2B.
[0108] Analysis system 300 may include a computing platform 302 and a data store 304. In some embodiments, analysis system 300 also includes a display system 306. Computing platform 302 can take a variety of forms. In one or more embodiments, computing platform 302 includes a single computer (or computer system) or multiple computers in communication with each other. In other examples, computing platform 302 takes the form of a cloud computing platform.
[0109] The data store 304 and the display system 306 may each be in communication with the computing platform 302. In some examples, the data store 304, the display system 306, or both may be considered part of the computing platform 302 or may be otherwise integrated with the computing platform 302. Thus, in some examples, the computing platform 302, the data store 304, and the display system 306 may be separate components that communicate with each other, while in other examples, some combination of these components may be integrated together. Communication between these different components may be implemented using any number of wired, wireless, or optical communication links, or a combination thereof.
[0110] The analysis system 300 includes, for example, a peptide structure analyzer 308, which may be implemented using hardware, software, firmware, or a combination thereof. In one or more embodiments, the peptide structure analyzer 308 is implemented using a computing platform 302.
[0111] The peptide structure analyzer 308 receives peptide structure data 310 for processing. The peptide structure data 310 may be, for example, peptide structure data output from the sample preparation and processing 106 of Figures 1, 2A, and 2B. Thus, the peptide structure data 310 may correspond to the set of peptide structures 122 identified for the biological sample 112, and thus may correspond to the biological sample 112.
[0112] The peptide structure data 310 can be transmitted as input to the peptide structure analyzer 308, retrieved from or accessed from the data store 304 or some other type of storage (e.g., cloud storage), or obtained in some other manner. In some cases, the peptide structure data 310 can be retrieved from the data store 304 in response to (e.g., directly or indirectly based on) receiving user input entered by a user via an input device.
[0113] The peptide structure data 310 may include quantification data for a plurality of peptide structures. For example, the peptide structure data 310 may include a set of quantification metrics for each peptide structure of the plurality of peptide structures. The quantification metric for a peptide structure may be selected as one of a relative amount, an adjusted amount, a normalized amount, a relative abundance, an adjusted abundance, and a normalized abundance. In some cases, the quantification metric for a peptide structure is selected from one of a relative concentration, an adjusted concentration, and a normalized concentration. In this manner, the peptide structure data 310 may provide abundance information for a plurality of peptide structures for the biological sample 112.
[0114] In some embodiments, a peptide structure in set of peptide structures 312 includes a glycosylated peptide structure or glycopeptide structure defined by a peptide sequence and a glycan structure attached to a linkage site of the peptide sequence. For example, a peptide structure may be a glycopeptide or a portion of a glycopeptide. In some embodiments, a peptide structure in set of peptide structures 312 includes a deglycosylated peptide structure defined by a peptide sequence. For example, a peptide structure may be a peptide or a portion of a peptide, which may be referred to as a quantification peptide.
[0115] The set of peptide structures 312 may be identified as most predictive of or most associated with a symptomatic disease state based on training of the model 314. In one or more embodiments, the set of peptide structures 312 includes at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, at least twelve, at least thirteen, at least fourteen, at least fifteen, at least sixteen, at least seventeen, at least eighteen, at least nineteen, at least twenty, at least twenty-one, at least twenty-two, at least twenty-three, at least twenty-four, at least twenty-six ... At least 25, at least 26, at least 27, at least 28, at least 29, at least 30, at least 31, at least 32, at least 33, at least 34, at least 35, at least 36, at least 37, at least 38, at least 39, at least 40, at least 41, at least 42, at least 43, at least 44, at least 45, at least 46, at least 47, at least 48, at least 49, at least 50, at least 51, at least 52, or all 53. The number of peptide structures selected from Table 1 for inclusion in set of peptide structures 312 may be based, for example, on the desired level of accuracy. In one or more embodiments, 40 peptide structures are selected from Table 1 for inclusion in set of peptide structures 312.
[0116] In Table 1, "PS-ID number" identifies a label or index for the peptide structure; "peptide structure (PS) name" identifies the name of the peptide structure; "protein sequence number" identifies the sequence ID of the protein associated with the peptide structure (e.g., from which the peptide structure is derived); "peptide sequence number" identifies the peptide sequence number for the peptide sequence of the peptide structure; "monoisotopic mass" identifies the monoisotopic mass of the peptide structure in Daltons (Da); "linkage site location in protein sequence" identifies the site location relative to the protein sequence to which the corresponding glycan structure is attached; "linkage site location in peptide sequence" identifies the site location relative to the peptide sequence to which the corresponding glycan structure is attached; and "GL number" identifies a label or index for the corresponding glycan structure. For glycopeptide structures, the name of the peptide structure includes an abbreviation for the protein associated with the peptide structure, a first number corresponding to the linkage site location relative to the protein sequence of the protein, and a second number identifying the glycan attached to the protein. In the case of deglycosylated peptide structures, the name of the peptide structure includes the abbreviation of the protein associated with the peptide structure and the corresponding peptide sequence of the peptide structure. [Table 1-1] [Table 1-2] [Table 1-3]
[0117] In one or more embodiments, the set of peptide structures 312 includes only peptide structures fragmented from alpha-2-macroglobulin (A2MG), and thus only A2MG glycoforms. In one or more embodiments, the set of peptide structures 312 includes only peptide structures fragmented from alpha-1-acid glycoprotein 1 (AGP1), and thus only AGP1 glycoforms. In one or more embodiments, the set of peptide structures 312 includes only peptide structures fragmented from haptoglobin (HPT), and thus only HPT glycoforms. In one or more embodiments, the set of peptide structures 312 includes only peptide structures fragmented from complement factor H (CFAH), and thus only CFAH glycoforms. In one or more embodiments, the set of peptide structures 312 includes only peptide structures fragmented from alpha-1-antitrypsin (A1AT), and thus only A1AT glycoforms. In some embodiments, the set of peptide structures 312 includes only peptide structures fragmented from at least one of A2MG, AGP1, HPT, CFAH, or A1AT.
[0118] The peptide structure analyzer 308 includes a model 314 configured to receive peptide structure data 310 for processing. The model 314 may be implemented in any of a number of different ways. The model 314 may be implemented using any number of models, functions, equations, algorithms, and / or other mathematical techniques.
[0119] In one or more embodiments, model 314 includes machine learning model 316, which may itself be composed of any number of machine learning models and / or algorithms. For example, machine learning model 316 may include, but is not limited to, at least one of a parametric model, a non-parametric model, a deep learning model, a neural network, a linear discriminant analysis model, a quadratic discriminant analysis model, a support vector machine, a random forest algorithm, a nearest neighbor algorithm (e.g., a k-nearest neighbor algorithm), a combined discriminant analysis model, a k-means clustering algorithm, an unsupervised model, a logistic regression model, a multivariate regression model, a penalized multivariate regression model, or another type of model. In various embodiments, model 314 includes machine learning model 316 that includes any number or combination of the above models or algorithms.
[0120] In various embodiments, the model 314 analyzes a portion (e.g., some or all) of the peptide structure data 310 corresponding to the set of peptide structures 312 to generate a disease indicator 318 that classifies the biological sample 112 as evidence of a corresponding state among a plurality of states 320 associated with FLD progression. The disease indicator 318 can take various forms. In one or more embodiments, the disease indicator 318 is a score that indicates a classification of the corresponding state for the biological sample 112. For example, each of the states 320 may be associated with a different range of score values. If the score falls within a selected range associated with a particular state among the states 320, the score indicates that the biological sample 112 evidences that particular state. Thus, the score provides a classification of the biological sample 112 as corresponding to that particular state.
[0121] In other embodiments, the disease indicator 318 includes a score indicating a probability that a subject (e.g., subject 114 of FIG. 1 ) will be in one of the conditions 320 associated with FLD progression. For example, the disease indicator 318 may include one or more scores, each of which may indicate whether the biological sample 112 will demonstrate a corresponding one of the conditions 320 associated with FLD progression. In some examples, the disease indicator 318 includes a score for each of the conditions 320 associated with FLD progression. A higher score indicates a higher probability that the biological sample 112 will demonstrate the corresponding condition.
[0122] In various embodiments, the machine learning model 316 takes the form of a regression model 320. The regression model 320 may include, for example, at least one lasso regression model (or lasso regularization model) that is trained to calculate the disease indicators 318. The regression model 320 may be trained to identify weight coefficients for peptide structures in the set of peptide structures 312.
[0123] The peptide structure analyzer 308 may generate the final output 128 based on the disease indicators 318 output by the model 314. In other embodiments, the final output 128 may be the output generated by the model 314.
[0124] In some embodiments, the final output 128 includes a disease indicator 318. In other embodiments, the final output 128 includes a diagnostic output 324 and / or a therapeutic output 326. The diagnostic output 324 may include, for example, an identification of a classification of which of the conditions 320 is evidenced by the biological sample 112 based on the disease indicator 318. The therapeutic output 326 may include, for example, at least one of identification of a therapeutic agent for treating the subject, design of a therapeutic agent, or a treatment regimen for administering a therapeutic agent. In some embodiments, the therapeutic agent is an immune checkpoint inhibitor.
[0125] In some examples, the final output 128 may be transmitted to a remote system 130 for processing. In other embodiments, the final output 128 may be displayed on a graphical user interface 328 of the display system 306 for viewing by a human operator. The human operator may use the final output 128 to diagnose and / or treat the subject when the final output 128 indicates that the subject is positive for a condition (e.g., NASH, HCC) that follows the disease progression of the disease (e.g., FLD).
[0126] VB computer implementation system 4 is a block diagram of a computer system according to various embodiments. Computer system 400 may be an example of one implementation for computing platform 302 described above in FIG.
[0127] In one or more examples, computer system 400 may include a bus 402 or other communication mechanism for communicating information, and a processor 404 coupled with bus 402 for processing information. In various embodiments, computer system 400 may also include memory, which may be random access memory (RAM) 406 or other dynamic storage device, coupled to bus 402 for determining instructions to be executed by processor 404. The memory may also be used for storing temporary variables or other intermediate information during execution of instructions to be executed by processor 404. In various embodiments, computer system 400 may further include read-only memory (ROM) 408 or other static storage device coupled to bus 402 for storing static information and instructions for processor 404. A storage device 410, such as a magnetic or optical disk, may be provided and coupled to bus 402 for storing information and instructions.
[0128] In various embodiments, computer system 400 may be coupled via bus 402 to a display 412, such as a cathode ray tube (CRT) or liquid crystal display (LCD), for displaying information to a computer user. An input device 414, including alphanumeric and other keys, may be coupled to bus 402 for communicating information and command selections to processor 404. Another type of user input device is a cursor control 416, such as a mouse, joystick, trackball, gesture input device, eye-gaze-based input device, or cursor direction keys, for communicating directional information and command selections to processor 404 and for controlling cursor movement on display 412. This input device 414 typically has two degrees of freedom in two axes, a first axis (e.g., x) and a second axis (e.g., y), allowing the device to specify a position in a plane. However, it should be understood that input devices 414 that allow cursor movement in three dimensions (e.g., x, y, and z) are also contemplated herein.
[0129] Consistent with certain practice of the present teachings, results may be provided by computer system 400 in response to processor 404 executing one or more sequences of one or more instructions contained in RAM 406. Such instructions may be read into RAM 406 from another computer-readable medium or computer-readable storage medium, such as storage device 410. Execution of the sequences of instructions contained in RAM 406 may cause processor 404 to perform the processes described herein. Alternatively, hard-wired circuitry may be used in place of or in combination with software instructions to practice the present teachings. Thus, practice of the present teachings is not limited to any specific combination of hardware circuitry and software.
[0130] The terms "computer-readable medium" (e.g., data store, data storage, storage device, data storage device, etc.) or "computer-readable storage medium," as used herein, refer to any medium that participates in providing instructions to processor 404 for execution. Such media may take many forms, including, but not limited to, non-volatile media, volatile media, and transmission media. Examples of non-volatile media may include, but are not limited to, optical disks, solid-state disks, and magnetic disks, such as storage device 410. Examples of volatile media may include, but are not limited to, dynamic memory, such as RAM 406. Examples of transmission media may include, but are not limited to, coaxial cables, copper wire, and fiber optics, including the wires that comprise bus 402.
[0131] Common forms of computer-readable media include, for example, floppy disks, flexible disks, hard disks, magnetic tape or any other magnetic medium, CD-ROMs, any other optical medium, punch cards, paper tape, any other physical medium with a pattern of holes, RAM, PROMs, and EPROMs, FLASH-EPROMs, any other memory chip or cartridge, or any other tangible medium from which a computer can read.
[0132] In addition to computer-readable media, instructions or data may be provided as signals on a transmission medium included in a communications device or system to provide sequences of one or more instructions to the processor 404 of the computer system 400 for execution. For example, a communications device may include a transceiver having signals indicative of instructions and data. The instructions and data are configured to cause one or more processors to implement the functions outlined herein in this disclosure. Representative examples of data communications transmission connections may include, but are not limited to, a telephone modem connection, a wide area network (WAN), a local area network (LAN), an infrared data connection, an NFC connection, an optical communications connection, etc.
[0133] It should be understood that the methodologies, flowcharts, diagrams, and accompanying disclosure described herein may be implemented using computer system 400 as a standalone device or over a distributed network of shared computer processing resources, such as a cloud computing network.
[0134] The methodologies described herein can be implemented by various means depending on the application. For example, these methodologies can be implemented in hardware, firmware, software, or any combination thereof. In the case of hardware implementation, the processing unit can be implemented in one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic circuits (PLDs), field-programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, electronic devices, other electronic units designed to perform the functions described herein, or combinations thereof.
[0135] In various embodiments, the methods of the present teachings may be implemented as firmware and / or software programs and applications written in conventional programming languages, such as C, C++, Python, etc. When implemented as firmware and / or software, the embodiments described herein may be implemented in a non-transitory computer-readable medium having stored thereon a program for causing a computer to perform the methods. It should be understood that the various engines described herein may be provided on a computer system, such as computer system 400, whereby processor 404 performs the analyses and decisions provided by these engines according to instructions provided by any one or combination of memory components RAM 406, ROM 408, or storage device 410, and user input provided via input device 414.
[0136] VI. Exemplary Methodologies for Diagnosis Based on Peptide Structure Data Analysis VI.A. General Methodology 5 is a flowchart of a process for classifying a biological sample obtained from a subject with respect to multiple conditions associated with fatty liver disease (FLD) progression. Process 500 may be practiced, for example, using at least a portion of workflow 100 described in FIGS. 1, 2A, and 2B and / or analysis system 300 described in FIG. 3. Process 500 may be used to generate a diagnostic output, such as diagnostic output 324 of FIG. 3.
[0137] Step 502 includes receiving peptide structure data corresponding to a set of glycoproteins in a biological sample obtained from a subject. The peptide structure data may be, for example, an example implementation of peptide structure data 310 in FIG. 3. The peptide structure data may include quantification data for each peptide structure of a plurality of peptide structures. The peptide structure data may be generated using, for example, but not limited to, multiple reaction monitoring mass spectrometry (MRM-MS). In one or more embodiments, the peptide structure data includes quantification data for a plurality of peptide structures. This quantification data for a peptide structure may include, for example, but not limited to, at least one of abundance, relative abundance, normalized abundance, relative amount, adjusted amount, normalized amount, relative concentration, adjusted concentration, or normalized concentration of the peptide structure. In one or more embodiments, the quantification data includes the normalized abundance of the peptide structure.
[0138] In one or more embodiments, peptide structure data is generated for a sample created from a biological sample. For example, the biological sample may be prepared using reduction, alkylation, and enzymatic digestion to form a prepared sample. The prepared sample includes a plurality of peptide structures for which peptide structure data is generated and then received in step 502.
[0139] Step 504 includes inputting quantification data identified from the peptide structure data for the set of peptide structures into a machine learning model. In some embodiments, the set of peptide structures includes at least one peptide structure identified from the plurality of peptide structures in Table 1 above. In various embodiments, the at least one peptide structure includes a glycopeptide structure defined by the peptide sequence and a glycan structure linked to the peptide sequence at a linkage site of the peptide sequence, as identified in Table 1.
[0140] The quantification data may include, for example, one or more quantification metrics for each peptide structure of the plurality of peptide structures. The quantification metric for a peptide structure may be, for example, but not limited to, a relative amount, an adjusted amount, a normalized amount, a relative concentration, an adjusted concentration, or a normalized concentration. In this manner, the quantification data for a given peptide structure indicates the abundance of the peptide structure in the biological sample. In one or more embodiments, the quantification data includes a normalized abundance of the peptide structure. The machine learning model may be, for example, a supervised machine learning model.
[0141] Step 506 includes analyzing the quantified data using a machine learning model to generate a disease indicator. In various embodiments, the generated disease indicator may include an indication of whether the biological sample evidences a condition associated with FLD progression. For example, the disease indicator may indicate whether the biological sample is likely to be positive for NASH or HCC, or whether it is likely to be healthy or non-NASH / HCC (e.g., may be indicative of a benign liver mass). In various embodiments, the disease indicator includes a probability that the biological sample evidences an NASH condition, or a probability that the biological sample evidences an HCC condition. In one or more embodiments, the machine learning model may generate an output that classifies the biological sample as positive for an NASH condition, positive for an HCC condition, or positive for a healthy (or non-NASH / HCC) condition. In various embodiments, the disease indicator may be a score.
[0142] Step 508 includes generating a diagnostic output based on the disease indicators classifying the biological sample as evidencing a corresponding state among a plurality of states associated with FLD progression. In various embodiments, the plurality of states may include a NASH state and a hepatocellular carcinoma (HCC) state. In some embodiments, the plurality of states may include a non-NASH / HCC state, including at least one of a healthy state, a benign liver mass state, a state without liver disease, or some other type of non-NASH / HCC state.
[0143] If the disease indicator is a score, step 508 may include determining that the score falls within a selected range associated with a corresponding one of the conditions (e.g., 0.0 to 0.05 for a non-NASH / HCC condition, 0.05 to 0.4 for a NASH condition, and 0.4 to 1.0 for a HCC condition). In some embodiments, step 506 may include determining that the biological sample evidences the corresponding condition based on a determination that the score falls within the selected range associated with the corresponding condition.
[0144] In one or more embodiments, generating a diagnostic output in step 508 includes generating the diagnostic output as part of a report identifying a corresponding condition. In some embodiments, step 508 may also include generating a therapeutic output based on at least one of the diagnostic output or the disease indicator. In some embodiments, the therapeutic output includes at least one of an identification of a therapeutic for treating the subject, a design for the therapeutic, a manufacturing plan for the therapeutic, or a therapeutic plan for administering the therapeutic.
[0145] For subjects diagnosed with NASH, treatment may include, for example, a therapeutic dose of at least one of obeticholic acid (OCA), tropifexor, elafibranor, saroglitazar, aramchol, semaglutide, tirzepatide, cotadutide, NGM282, MSDC-0602K, resmetirom, cenicriviroc, selonsertib, emricasan, simtuzumab, and GR-MD-02. For subjects diagnosed with NASH, treatment may include, for example, a therapeutic dose of at least one of atezolizumab, bevacizumab, sorafenib, lenvatinib, nivolumab, regorafenib, cabozantinib, pemigatinib, ramucirumab, or pembrolizumab.
[0146] 6 is a flowchart of a process for detecting the presence of one of a plurality of conditions associated with fatty liver disease (FLD) progression in a biological sample. Process 600 may be practiced, for example, using at least a portion of workflow 100 described in FIGS. 1, 2A, and 2B and / or analysis system 300 described in FIG. 3.
[0147] Step 602 includes receiving peptide structure data corresponding to a set of glycoproteins in a biological sample obtained from a subject. The peptide structure data may be generated from a prepared sample using, for example, multiple reaction monitoring mass spectrometry (MRM-MS). The peptide structure data may include quantification data for each peptide structure in the panel of peptide structures. The quantification data for a peptide structure in the plurality of peptide structures may include at least one of a relative amount, an adjusted amount, a normalized amount, a relative concentration, an adjusted concentration, a normalized concentration, or another quantification metric.
[0148] Step 604 includes analyzing the peptide structure data using a supervised machine learning model to generate disease indicators based on at least three peptide structures selected from the group of peptide structures identified in Table 1. In one or more embodiments, the supervised machine learning model includes a logistic regression model.
[0149] In various embodiments, the at least three peptide structures comprise a glycopeptide structure defined by a peptide sequence identified in Table 1 and a glycan structure linked to the peptide sequence at a linkage site of the peptide sequence, wherein the peptide sequence is one of SEQ ID NOs: 23, 24, 25, 29, 30, 31, 33, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, or 52 as defined in Table 1.
[0150] In one or more embodiments, the at least three peptide structures comprise only peptide structures fragmented from alpha-2-macroglobulin (A2MG). In one or more embodiments, the at least three peptide structures comprise only peptide structures fragmented from alpha-1-acid glycoprotein 1 (AGP1). In one or more embodiments, the at least three peptide structures comprise only peptide structures fragmented from haptoglobin (HPT). In one or more embodiments, the at least three peptide structures comprise only peptide structures fragmented from complement factor H (CFAH). In one or more embodiments, the at least three peptide structures comprise only peptide structures fragmented from alpha-1-antitrypsin (A1AT).
[0151] In various embodiments, step 604 may include calculating a peptide structure profile for the biological sample, identifying a weighted value for each peptide structure of the at least three peptide structures. In some embodiments, the weighted value for a peptide structure of the at least three peptide structures may be a product of a quantification metric for the peptide structure identified from the peptide structure data and a weighting coefficient for the peptide structure. The weighting coefficient for a corresponding peptide structure of the at least three peptide structures may indicate the relative significance of the corresponding peptide structure for the disease indicator. In one or more embodiments, analyzing the peptide structure data includes calculating a disease indicator using the peptide structure profile. The disease indicator may be, for example, a score indicating which of a plurality of conditions is evidenced by the biological sample.
[0152] In one or more embodiments, the supervised machine learning model uses three different models in generating the final disease indicator. Each of these three models may be, for example, a regression model used to analyze peptide structure quantification data for a particular condition versus other conditions. For example, each of these three regression models may be used to distinguish a given condition from other conditions in the plurality of conditions.
[0153] In one or more embodiments, the at least three peptide structures of step 604 may be selected from the peptide structures identified in Table 2 below, which includes a subset of the peptide structures identified in Table 1. Table 2 further identifies weighting factors associated with distinguishing between a NASH state and other states (e.g., an HCC state and a control state), an HCC state and other states (e.g., an NASH state and a control state), and a control state and other states (e.g., an NASH state and an HCC state).
[0154] In Table 2, "PS-ID Number" identifies the label or index of the peptide structure, "Peptide Structure (PS) Name" identifies the name of the peptide structure, "Protein SEQ ID NO" identifies the sequence ID of the protein associated with the peptide structure (e.g., from which the peptide structure is derived), "Peptide SEQ ID NO" identifies the peptide SEQ ID NO for the peptide sequence of the peptide structure, "Coefficient (Control vs. Rest)" identifies the weighting coefficient between the control state and other states, "Coefficient (NASH vs. Rest)" identifies the weighting coefficient between NASH and other states, and "Coefficient (HCC vs. Rest)" identifies the weighting coefficient between HCC and other states. [Table 2-1] [Table 2-2]
[0155] Step 606 includes detecting the presence of a corresponding condition among a plurality of conditions associated with FLD progression in response to determining that the disease indicator falls within a selected range associated with the corresponding condition. In some embodiments, the plurality of conditions includes at least two selected from the group consisting of a non-alcoholic steatohepatitis (NASH) condition, a hepatocellular carcinoma (HCC) condition, or a non-NASH / HCC condition (e.g., a non-alcoholic fatty liver disease condition, a control condition, a healthy condition, a benign liver tumor condition, or a condition without liver disease).
[0156] In one or more embodiments, the corresponding condition is a non-NASH / HCC condition, and the selected range of disease indicators associated with the non-NASH / HCC condition is 0.00 to about 0.05. In some embodiments, the corresponding condition is a NASH condition, and the selected range of disease indicators associated with the NASH condition is 0.05 to 0.4. In some embodiments, the corresponding condition is an HCC condition, and the selected range of disease indicators associated with the HCC condition is 0.4 to 1.0.
[0157] In one or more embodiments, the process 600 may further include generating a report including a diagnosis based on the corresponding condition detected for the subject. The report may include, for example, a disease indicator.
[0158] VI.B. Model Training 7 is a flowchart of a process for training a model to generate a classification of a biological sample as belonging to or evidence of one of a plurality of states associated with FLD progression. Process 700 may be implemented using, for example, analysis system 300 described in FIG. 3. In one or more embodiments, process 700 may be performed to train model 314 of FIG. 3.
[0159] Step 702 includes receiving quantification data for a panel of peptide structures for a plurality of subjects diagnosed with a plurality of conditions associated with FLD progression, the quantification data including a plurality of peptide structure profiles for the plurality of subjects, and identifying a corresponding state among the plurality of conditions for each peptide structure profile of the plurality of peptide structure profiles.
[0160] In one or more embodiments, the plurality of peptide structure profiles comprises normalized abundance data for a first number of peptide structures. In one or more embodiments, the quantification data of the panel of peptide structures for the plurality of subjects diagnosed with the plurality of conditions comprises at least one of abundance, relative abundance, normalized abundance, relative amount, adjusted amount, normalized amount, relative concentration, adjusted concentration, or normalized concentration.
[0161] Step 704 includes training a supervised machine learning model using the quantified data to determine which of the plurality of states the biological sample from the subject corresponds to. In various embodiments, the supervised machine learning model includes a logistic regression model. In one or more embodiments, the logistic regression model may be a Lasso regularization model. The plurality of states may include, for example, at least one of a NASH state, an HCC state, or a non-NASH / HCC state.
[0162] In one or more embodiments, step 704 may include training a supervised machine learning model to make a state determination based on quantification data for at least three peptide structures, for example, from the list of peptide structures identified in Table 1 or Table 2 above.
[0163] The peptide structures selected for training may be, for example, peptide structures identified from a differential expression analysis performed using initial training data to compare quantification data for peptide structures from a first portion of a plurality of subjects diagnosed with NASH, a second portion of a plurality of subjects diagnosed with HCC, and a third portion of a plurality of subjects diagnosed with a non-NASH / HCC condition (e.g., a healthy condition, a benign liver mass, a condition without liver disease, etc.).
[0164] For example, comparisons may be made to compare peptide structure quantification data between a first portion of subjects with NASH and a third portion of subjects with a non-NASH / HCC condition (e.g., a control condition), to compare a second portion of subjects with HCC with the first portion of subjects with NASH, and / or to compare the first portion of subjects with NASH with the second portion of subjects with HCC. In some embodiments, comparisons may be normalized and compared to each other.
[0165] Tables 3A-3C below show the fold change (FC), false discovery rate (FDR), and p-value calculated based on such comparisons. [Table 3-1] [Table 3-2] [Table 4-1] [Table 4-2] [Table 5-1] [Table 5-2]
[0166] VI.C. Combined Training and Diagnostics The exemplary methodology described in Section VI can be used to diagnose subjects who may be suffering from FLD. This diagnosis can be used to determine a treatment regimen for the subject. The embodiments described herein may enable faster and more accurate diagnosis of NASH vs. HCC status. The ability to more quickly and accurately diagnose subjects (or patients) who have progressed from NASH to HCC during the progression of FLD may allow for faster treatment of the subject, which may result in a more favorable treatment outcome for the subject. Furthermore, being able to more quickly and accurately determine when a subject has progressed from NASH to HCC may be particularly useful in reducing the need for hospitalization and avoiding death.
[0167] 8 is a flowchart of a process for classifying a biological sample as corresponding to one of a plurality of conditions associated with fatty liver disease (FLD) progression, according to one or more embodiments. Process 800 may be practiced using at least a portion of workflow 100 described in FIGS. 1, 2A, and / or 2B and / or analysis system 300 described in FIG. 3.
[0168] Step 802 includes training a supervised machine learning model using training data. In one or more embodiments, the training data includes a plurality of peptide structure profiles of a plurality of training subjects, and identifies a corresponding state of a plurality of states for each peptide structure profile of the plurality of peptide structure profiles.
[0169] Step 804 includes receiving peptide structure data corresponding to a set of glycoproteins in a biological sample obtained from a subject.
[0170] Step 806 includes inputting the quantification data identified from the peptide structure data for the set of peptide structures into a trained supervised machine learning model. In some embodiments, the set of peptide structures includes at least one peptide structure identified in Table 1.
[0171] Step 808 includes analyzing the quantified data using a supervised machine learning model to generate a score.
[0172] Step 810 includes determining that the score falls within a selected range associated with a corresponding state of a plurality of states associated with FLD progression.
[0173] Step 812 includes generating a diagnostic output indicating that the biological sample evidences a corresponding condition. In some embodiments, the conditions include a non-alcoholic steatohepatitis (NASH) condition, a hepatocellular carcinoma (HCC) condition, and a non-NASH / HCC (e.g., control or healthy) condition.
[0174] VII. Exemplary Methodologies for Treatment VII.A. Treatment of NASH 9 is a flowchart of a process for treating a subject for NASH, according to one or more embodiments. Process 900 may be practiced, at least in part, using workflow 100 described in FIGS. 1, 2A, and / or 2B and / or at least a portion of analysis system 300 described in FIG. 3.
[0175] Step 902 includes receiving a biological sample, which may be obtained from a patient.
[0176] Step 904 involves using an MRM-MS system to determine the abundance of each peptide structure identified in the predetermined list, which may be, for example, the list identified in Table 1 or the list identified in Table 2.
[0177] Step 906 involves using a machine learning model to analyze the abundance of each peptide structure to generate a disease indicator.
[0178] Step 908 includes generating a diagnostic output based on the disease indicators that classify the biological sample as evidence that the patient has NASH (or a NASH disorder).
[0179] Step 910 includes administering a therapeutic agent for NASH to the patient via at least one of intravenous and oral administration of the therapeutic agent at a therapeutic dose(s). The treatment may consist of one or more therapeutic agents or derivatives thereof.
[0180] Therapeutic agents include, but are not limited to, at least one compound or derivative thereof selected from the group consisting of obeticholic acid (OCA), tropifexor, elafibranor, saroglitazar, aramchol, semaglutide, tirzepatide, cotadutide, NGM282, MSDC-0602K, resmetirom, cenicriviroc, selonsertib, emricasan, simtuzumab, and GR-MD-02. In one or more embodiments, the therapeutic dose of obeticholic acid (OCA) may include a dose in the range of 10 to 25 mg per day.
[0181] Process 900 may include one or more additional steps. For example, process 900 may further include designing a therapeutic agent for treating the subject in response to determining that the biological sample obtained from the subject is evidence of NASH. Process 900 may also include generating a treatment plan for treating the subject in response to determining that the biological sample obtained from the subject is evidence of NASH.
[0182] VII.B. Treatment of HCC 10 is a flowchart of a process for treating a subject for HCC according to one or more embodiments. Process 1000 may be practiced, at least in part, using workflow 100 described in FIGS. 1, 2A, and / or 2B and / or at least a portion of analysis system 300 described in FIG. 3.
[0183] Step 1002 includes receiving a biological sample, which may be obtained from a patient.
[0184] Step 1004 involves using an MRM-MS system to determine the abundance of each peptide structure identified in the predetermined list, which may be, for example, the list identified in Table 1 or the list identified in Table 2.
[0185] Step 1006 involves using a machine learning model to analyze the abundance of each peptide structure to generate a disease indicator.
[0186] Step 1008 includes generating a diagnostic output based on the disease indicators that classify the biological sample as evidence that the patient has an HCC disorder.
[0187] Step 1010 includes administering a therapeutic agent for HCC to the patient via at least one of intravenous and oral administration of the therapeutic agent at a therapeutic dose(s). The treatment may consist of one or more therapeutic agents or derivatives thereof.
[0188] The therapeutic agent may include, but is not limited to, at least one compound or derivative thereof selected from the group consisting of atezolizumab, bevacizumab, sorafenib (e.g., sorafenib tosylate), lenvatinib (e.g., lenvatinib mesylate), nivolumab, regorafenib, cabozantinib (e.g., cabozantinib-S-malate), pemigatinib, ramucirumab, or pembrolizumab. In one or more embodiments, step 1010 includes administering sorafenib or a derivative thereof to the patient via at least one of intravenous or oral administration in the range of 775 to 825 mg daily. In one or more embodiments, step 1010 includes administering lenvatinib or a derivative thereof to the patient via at least one of intravenous or oral administration in the range of 7.5-8.5 mg / day when the patient weighs less than 60 kg and 11.5-12.5 mg / day when the patient weighs more than 60 kg. In one or more embodiments, step 1010 includes administering nivolumab or a derivative thereof to the patient via at least one of intravenous or oral administration in the range of 0.75-1.25 mg / kg. In one or more embodiments, step 1010 includes administering regorafenib or a derivative thereof to the patient via oral administration in the range of 150-170 mg / day. In one or more embodiments, step 1010 includes administering cabozantinib or a derivative thereof to the patient via at least one of intravenous or oral administration in the range of 50-70 mg / day. In one or more embodiments, step 1010 includes administering ramucirumab or a derivative thereof to the patient via at least one of intravenous or oral administration in the range of 8-12 mg / kg.
[0189] Process 1000 may include one or more additional steps. For example, process 1000 may further include designing a therapeutic agent for treating the subject in response to determining that the biological sample obtained from the subject evidences HCC. Process 900 may include generating a treatment plan for treating the subject in response to determining that the biological sample obtained from the subject evidences HCC.
[0190] VIII. PEPTIDE STRUCTURES AND PRODUCT ION COMPOSITIONS, KITS, AND REAGENTS Aspects of the present disclosure include compositions comprising one or more of the peptide structures listed in Table 1. In some embodiments, a composition comprises a plurality of the peptide structures listed in Table 1. In some embodiments, the composition comprises 1, 2, 3, 4, 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, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, or all 53 of the peptide structures listed in Table 1. In some embodiments, the composition comprises a peptide structure having an amino acid sequence with at least 80% sequence identity, such as at least 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100%, to any one of SEQ ID NOs: 23-52 listed in Table 1.
[0191] Embodiments of the present disclosure include compositions comprising one or more precursor ions having a defined charge and / or a defined mass-to-charge (m / z) ratio, as listed in Table 4 below. Embodiments of the present disclosure include compositions comprising one or more product ions having a defined mass-to-charge (m / z) ratio, which are produced by converting a peptide structure described herein (e.g., a peptide structure listed in Table 1) into gas phase ions in a mass spectrometry system. Conversion of the peptide structure into gas phase ions can be performed using any of a variety of techniques, including, but not limited to, matrix-assisted laser desorption ionization (MALDI), electron ionization (EI), electrospray ionization (ESI), atmospheric pressure chemical ionization (APCI), and / or atmospheric pressure photoionization (APPI).
[0192] Aspects of the present disclosure include compositions comprising one or more product ions produced from one or more of the peptide structures described herein (e.g., the peptide structures listed in Table 1). In some embodiments, the composition comprises a set of product ions listed in Table 4 having m / z ratios selected from the list provided for each peptide structure in Table 4.
[0193] In some embodiments, the composition comprises at least one of the peptide structures PS-1 through PS-53 identified in Table 1.
[0194] In some embodiments, the composition comprises a peptide structure or product ion comprising an amino acid sequence having at least 90% sequence identity to any one of SEQ ID NOs: 23-52 identified in Table 5 below, which correspond to peptide structures PS-1 through PS-53 in Table 1.
[0195] In some embodiments, the product ions are selected as one from the group consisting of product ions identified in Table 4, including product ions that fall within a specified m / z range of m / z ratios identified in Table 4 and are characterized by having precursor ions with m / z ratios within a specified m / z range of m / z ratios identified in Table 4. The first range of product ion m / z ratios may be ±0.5. The second range of product ion m / z ratios may be ±0.8. The third range of product ion m / z ratios may be ±1.0. The first range of precursor ion m / z ratios may be ±1.0 and the second range of precursor ion m / z ratios may be (±1.5). Thus, a composition may include product ions having m / z ratios within at least one of the first (±0.5), second (±0.8), or third ranges of product ion m / z ratios identified in Table 4, and may be characterized as having precursor ions having m / z ratios within at least one of the first (±0.5), second (±1.0), or third ranges (±1.0) of precursor ion m / z ratios identified in Table 4.
[0196] In Table 4, "PS-ID Number" identifies the peptide structure label or index; "R; "RT (min)" identifies the minimum retention time; "Collision Energy (Coll. Energy)" identifies the collision energy; "Precursor m / z" identifies the precursor ion m / z ratio; "Precursor Charge" identifies the precursor ion charge; "1st Product m / z" identifies the first product ion m / z ratio; "1st Product Charge" identifies the first product ion charge; "2nd Product Ion m / z" identifies the second product ion m / z ratio; and "2nd Product Charge" identifies the second product ion charge. [Table 6-1] [Table 6-2] [Table 6-3]
[0197] Table 5 defines the peptide sequences for SEQ ID NOs: 23-52 from Table 1. Table 5 further identifies the corresponding protein SEQ ID NO for each peptide sequence. The corresponding protein SEQ ID NO identifies the protein from which the peptide sequence may be derived. [Table 7]
[0198] Table 6 identifies the proteins of SEQ ID NOs: 1-22 from Table 1. Table 6 identifies the corresponding protein abbreviation and protein name for each of protein SEQ ID NOs: 1-22. Additionally, Table 6 identifies the corresponding Uniprot ID for each of protein SEQ ID NOs: 1-22. [Table 8]
[0199] Table 7 identifies and defines the glycan structures contained in Table 1. Table 7 identifies a graphical representation of one or more glycan structures associated with a particular glycan, and a coded representation of the composition for each glycan structure contained in Table 1. As used herein, the four-digit GL number is a designation representing the number of hexoses, the number of HexNAcs, the number of fucoses, and the number of neuraminic acids. [Table 9-1] [Table 9-2] [Table 9-3] [Table 9-4] [Table 9-5] [Table 9-6] [Table 9-7] [Table 9-8] [Table 10]
[0200] Aspects of the present disclosure include kits containing one or more compositions, each containing one or more peptide structures of the present disclosure that can be used as assay standards, and instructions for use. Kits according to one or more embodiments described herein may include a label indicating the intended use of the contents of the kit. The term "label," as used herein with respect to a kit, includes any writing or recorded material supplied on or with the kit, or otherwise accompanying the kit.
[0201] As described herein, peptide structures and the transition products produced therefrom may be useful for the diagnosis and treatment of various disease states within FLD, including, but not limited to, NASH and HCC. The transition products include a precursor ion and at least one product ion group. As described herein, the peptide structures in Table 1 and their corresponding precursor ions and product ion groups (ions having a defined m / z ratio or m / z ratios falling within the m / z ranges specified herein) can be used in mass spectrometry-based analyses to diagnose and facilitate treatment of diseases such as, for example, NASH and HCC.
[0202] Aspects of the present disclosure include methods for analyzing one or more peptide structures, as described herein. In some embodiments, the methods involve processing a sample from a patient, for example, using a liquid chromatography system (e.g., a high performance liquid chromatography system (HPLC)) to generate a prepared sample that can be input into a mass spectrometry system (e.g., a reaction monitoring mass spectrometry system). In certain embodiments, processing the sample can include performing one or more of a denaturation procedure, a reduction procedure, an alkylation procedure, and a digestion procedure. The denaturation and reduction procedure can be performed, for example, in a manner similar to denaturation and reduction 202 of FIG. 2. The alkylation procedure can be performed, for example, in a manner similar to alkylation procedure 204 of FIG. 2. The digestion procedure can be performed, for example, in a manner similar to digestion procedure 206 of FIG. 2.
[0203] In some embodiments, a method for analyzing one or more peptide structures involves detecting a set of product ions generated by a mass spectrometry system (e.g., a reaction monitoring mass spectrometry system), where one or more product ions may correspond to each of one or more peptide structures input to the mass spectrometry system. As described herein, each peptide structure may be converted to a set of product ions having a defined m / z ratio provided in Table 4, or an m / z ratio within a specified range of the m / z ratios provided in Table 4. In some embodiments, the method involves generating quantification (e.g., abundance) data for the one or more product ions detected using the mass spectrometry system.
[0204] In some embodiments, the method further includes generating a diagnostic output using the quantification data and a model trained using supervised and / or unsupervised machine learning. In certain embodiments, the reaction monitoring mass spectrometry system may include multiple / selected reaction monitoring mass spectrometry (MRM / SRM-MS) to detect one or more product ions and generate the quantification data.
[0205] IX. Representative Experimental Results IX.A. Sample Preparation and Mass Spectrometry Data Generation 11 is a table of sample populations used in experiments according to one or more embodiments. Samples used to generate the experimental results included serum samples from 23 patients with a biopsy-proven diagnosis of NASH (10 men, 13 women; Indivumed AG, Hamburg, Germany), 20 patients with a diagnosis of HCC (16 men, 4 women; 6 stage I, 8 stage II, 6 stage III, 2 stage IV; Indivumed AG), and 56 healthy subjects (controls, 26 men, 30 women) with no history of liver disease obtained from iSpecimen, Palleon Pharmaceuticals Inc. and the Human Immune Monitoring Center (HIMC) of Stanford. Clinical diagnoses of NASH and HCC patients were based on histopathological characterization of liver tissue obtained by needle biopsy or at the time of surgery.
[0206] 12 is a table of the sample population used for validation, according to one or more embodiments. The validation sample population consisted of serum samples from 28 control subjects (16 men, 12 women) diagnosed with benign liver masses and 28 subjects (20 men, 8 women) diagnosed with HCC, all obtained from Indivumed AG. Patient clinical diagnoses were based on histopathological characterization of liver tissue obtained by needle biopsy or at the time of surgery.
[0207] Prior to analysis, serum samples were reduced with DTT, alkylated with IAA, and then digested with trypsin in a water bath at 37°C for 18 hours. To quench the digestion, formic acid was added to each sample after incubation to a final concentration of 1%.
[0208] The digested serum samples were injected into a triple quadrupole mass spectrometer (MS) using a liquid chromatography system (e.g., a high-performance liquid chromatography (HPLC) system). Separation of peptide structures (glycosylated and deglycosylated) was performed using a 70-minute binary gradient. The triple quadrupole MS was operated in dynamic multiple reaction monitoring (dMRM) mode. Samples were injected in a randomized manner for the base phenotype, with a reference pooled serum digest interspersed with test samples at every 10th sample position throughout the run.
[0209] MRM analysis of the peptide structures revealed a total of 73 high-abundance serum glycoproteins. The transition list consisted of glycopeptide structures and deglycosylated peptide structures from each glycoprotein. The Python library Scikit-learn (https: / / scikit-learn.org / stable / ) was used for statistical analysis and building machine learning models.
[0210] The following formula was used to generate normalized abundance data for peptide structures: Normalized abundance = (raw abundance of peptide structure in sample / raw abundance of corresponding deglycosylated peptide from the same glycoprotein) / average relative abundance of the same glycopeptide or peptide in adjacent pooled reference serum samples
[0211] Relative abundance was calculated as the ratio of the raw abundance of any given glycopeptide to the sum of the raw abundances of all glycopeptides.
[0212] IX.B. Analysis of Peptide Structural Data (e.g., Abundance Data) and Validation of Their Diagnostic Power 13 is a graphical illustration of a plot of the mean normalized abundance of selected peptide structures identified and quantified via mass spectrometry, as shown, in accordance with one or more embodiments. As shown, the change in mean normalized abundance of the peptide structures increased or decreased in one direction as they progressed from a control state to a NASH state to an HCC state. This abundance data demonstrates that such peptide structures can be used to distinguish between control, NASH, and HCC states.
[0213] FIG. 14 is an illustration of a plot of normalized abundance of selected peptide structures of A2MG, according to one or more embodiments.
[0214] FIG. 15 is an illustration of a plot of normalized abundance of selected peptide structures of AGP1, according to one or more embodiments.
[0215] FIG. 16 is an illustration of a plot of normalized abundance of selected peptide structures of HPT, according to one or more embodiments.
[0216] FIG. 17 is an illustration of a plot of normalized abundance of selected peptide structures of CFAH, according to one or more embodiments.
[0217] FIG. 18 is an illustration of a plot of normalized abundance of selected peptide structures of A1AT, according to one or more embodiments.
[0218] Figures 14-18 show that the peptide structures of single glycoproteins (e.g., A2MG, AGP1, HPT, CFAH, A1AT) and combinations of such glycoproteins may be useful for diagnosing when a subject progresses from non-NASH / HCC (e.g., healthy) to NASH or from NASH to HCC.
[0219] A first regression model (e.g., Model 1) was trained to generate a disease indicator for a subject using the normalized abundance of various peptide structures (e.g., the peptide structures identified in Table 2). The disease indicator is generated as a score (e.g., a probability score), where the range within which the score falls allows for diagnosis or classification as a non-NASH / HCC state (e.g., a control state), a NASH state, or an HCC state.
[0220] IX.C. Verification 19 is a plot illustrating validation of the ability of disease indicators to distinguish between control, NASH, and HCC states, according to one or more embodiments. As shown, disease indicators of about 0.0 to about 0.05 were generally accurate in classifying as control states, disease indicators of about 0.05 to about 0.4 were generally accurate in classifying as NASH states, and disease indicators of about 0.5 to about 1.00 were generally accurate in classifying as HCC states.
[0221] Figure 20 is a plot of receiver operating characteristic (ROC) curves for distinguishing between NASH and HCC states for both the training and test sets, according to one or more embodiments. Leave-one-out cross validation (LOOCV) was performed on the normalized abundances of samples from both NASH and HCC patients. A logistic regression model with (lasso) normalization was iteratively trained on all samples, except for one left-out sample in each iteration. The trained model was then used to predict the left-out sample. As shown in Figure 20, the area under the ROC curve (AUROC) for the training set was found to be 0.99, while the AUROC for the test set was found to be 0.89.
[0222] IX.C.1. Validation using a second sample population To validate the ability of peptide structures to distinguish HCC from other conditions, a second regression model (Model 2) was used to analyze a second sample population (see Figure 12). Model 2 was trained using a smaller set of peptide structures to distinguish HCC from benign liver masses. For example, subjects used as controls were individuals diagnosed with benign liver masses. This allowed for a direct assessment of the discriminatory power of differential peptide structure abundance for HCC.
[0223] Ten peptide structures (eight glycopeptide structures and two peptide structures) were analyzed. All ten peptide structures were associated with A2MG. Model 2 was constructed using least absolute shrinkage and selection operator (LASSO) normalization and LOOCV.
[0224] FIG. 21 is a plot of the normalized abundance of these 10 peptide structures for control (eg, healthy) conditions versus HCC, and benign liver masses versus HCC, according to one or more embodiments.
[0225] 22 is a plot of receiver operating characteristic (ROC) curves for distinguishing HCC from benign liver masses for both the training set and the test set, according to one or more embodiments. The area under the ROC curve (AUROC) for the training set was found to be 0.85, while the AUROC for the test set was found to be 0.77. The AUROC for the discovery set (e.g., the same set used for Model 1) was found to be 0.85.
[0226] Table 8 provides the weighting coefficients of the trained Model 2 and AUC for each of the 10 peptide structures. [Table 11]
[0227] IX.C.2. Validation using three representative patients Three representative patients were tested using a trained regression model trained to distinguish HCC from one or more other conditions. Table 9 provides the normalized abundances determined for each patient for various peptide structures. The bottom of Table 9 provides the disease indicators calculated for these patients based on the normalized abundances. The disease indicators are probability scores indicating the likelihood that a subject has HCC. Using the disease indicators, only patient 3 was classified as having HCC, which was the correct diagnosis. [Table 12-1] [Table 12-2]
[0228] X. DESCRIPTION OF EMBODIMENTS Embodiment 1. A method for classifying a biological sample with respect to a plurality of conditions associated with fatty liver disease (FLD) progression, the method comprising: receiving peptide structure data corresponding to a set of glycoproteins in the biological sample obtained from a subject; inputting quantification data identified from the peptide structure data for the set of peptide structures into a machine learning model, wherein the set of peptide structures comprises at least one peptide structure identified from a plurality of peptide structures in Table 1; analyzing the quantification data using the machine learning model to generate a disease indicator; and generating a diagnostic output based on the disease indicator that classifies the biological sample as evidencing a corresponding condition among the plurality of conditions associated with FLD progression.
[0229] Embodiment 2. The method of embodiment 1, wherein the disease indicator comprises a score, and generating the diagnostic output comprises determining that the score falls within a selected range associated with the corresponding condition of the plurality of conditions, and determining that the biological sample evidences the corresponding condition in response to determining that the score falls within the selected range associated with the corresponding condition.
[0230] Embodiment 3. The method of embodiment 1 or embodiment 2, wherein the plurality of conditions comprises a nonalcoholic steatohepatitis (NASH) condition and a hepatocellular carcinoma (HCC) condition.
[0231] Embodiment 4. The method of any one of embodiments 1-3, wherein the plurality of conditions comprises non-NASH / HCC conditions, including at least one of a healthy condition, a condition without liver disease, or a benign liver mass condition.
[0232] Embodiment 5. The method of any one of embodiments 1-4, wherein the at least one peptide structure comprises a glycopeptide structure defined by a peptide sequence identified in Table 1 and a glycan structure linked to the peptide sequence at a linkage site of the peptide sequence, wherein the peptide sequence is one of SEQ ID NOs: 23, 24, 25, 29, 30, 31, 33, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, or 52 as defined in Table 1.
[0233] Embodiment 6. The method of any one of embodiments 1 to 5, further comprising training the machine learning model using training data, the training data comprising a plurality of peptide structure profiles for a plurality of subjects, and identifying a diagnosed condition among the plurality of conditions for each peptide structure profile of the plurality of peptide structure profiles.
[0234] Embodiment 7. The method of embodiment 6, further comprising: (1) performing a differential expression analysis using initial training data to compare at least one of a first portion of the plurality of subjects diagnosed with NASH to a second portion of the plurality of subjects diagnosed with HCC, the first portion of the plurality of subjects diagnosed with NASH to a third portion of the plurality of subjects assigned to a control state, or the second portion of the plurality of subjects diagnosed with HCC to a third portion of the plurality of subjects assigned to a control state; (2) identifying a group of peptide structures based on the differential expression analysis for use as a prognostic marker for the progression of FLD, wherein the group of peptide structures is identified in Table 1; and (3) forming the training data based on the identified group of peptide structures.
[0235] Embodiment 8. The method of embodiment 7, wherein the control condition comprises at least one of a healthy condition, a condition without liver disease, or a benign liver mass condition.
[0236] Embodiment 9. The method of any one of embodiments 1 to 8, wherein the machine learning model comprises a logistic regression model.
[0237] Embodiment 10. The method of any one of embodiments 1 to 9, wherein the quantification data for a peptide structure in the set of peptide structures comprises at least one of abundance, relative abundance, normalized abundance, relative amount, adjusted amount, normalized amount, relative concentration, adjusted concentration, or normalized concentration.
[0238] Embodiment 11. The method of any one of embodiments 1 to 9, wherein the quantification data for the set of peptide structures comprises a normalized abundance of each peptide structure in the set of peptide structures.
[0239] Embodiment 12. The method of any one of embodiments 1 to 11, wherein the peptide structural data is generated using multiple reaction monitoring mass spectrometry (MRM-MS).
[0240] Embodiment 13. The method of any one of embodiments 1 to 11, wherein the peptide structural data is generated using a liquid chromatography / mass spectrometry (LC / MS) system.
[0241] Embodiment 14. The method of any one of embodiments 1 to 13, further comprising creating a sample from the biological sample and preparing the sample using reduction, alkylation, and enzymatic digestion to form a prepared sample comprising the set of peptide structures.
[0242] Embodiment 15. The method of embodiment 14, further comprising generating said peptide structural data from said prepared samples using multiple reaction monitoring mass spectrometry (MRM-MS).
[0243] Embodiment 16. The method of any one of embodiments 1 to 15, wherein the biological sample comprises at least one of blood, serum, or plasma.
[0244] Embodiment 17. The method of any one of embodiments 1 to 16, wherein generating the diagnostic output includes generating a report identifying the corresponding condition.
[0245] Embodiment 18. The method of any one of embodiments 1 to 17, further comprising generating a treatment output based on at least one of the diagnostic output or the disease indicator.
[0246] Embodiment 19. The method of embodiment 18, wherein the treatment output includes at least one of an identification of a treatment for treating the subject, a design for the treatment, a manufacturing plan for the treatment, or a treatment plan for administering the treatment.
[0247] Embodiment 20. The method of embodiment 19, wherein the corresponding condition is a NASH condition and the treatment comprises at least one of obeticholic acid (OCA), tropifexor, elafibranor, saroglitazar, aramchol, semaglutide, tirzepatide, cotadutide, NGM282, MSDC-0602K, resmetirom, cenicriviroc, selonsertib, emricasan, simtuzumab, or GR-MD-02.
[0248] Embodiment 21. The method of embodiment 19, wherein the corresponding condition is an HCC condition and the treatment comprises at least one of atezolizumab, bevacizumab, sorafenib, lenvatinib, nivolumab, regorafenib, cabozantinib, pemigatinib, ramucirumab, or pembrolizumab.
[0249] Embodiment 22. The method of any one of embodiments 1 to 18, wherein the corresponding condition is nonalcoholic steatohepatitis (NASH), further comprising generating a treatment plan for the subject, wherein the treatment plan identifies at least one of a set of lifestyle modifications, a set of dietary modifications, coffee, vitamin E, calorie reduction, reduced salt intake, reduced sugar intake, or a set of cholesterol-reducing medications.
[0250] Embodiment 23. The method of any one of embodiments 1 to 18, wherein the corresponding condition is a NASH condition and the diagnostic output identifies the biological sample as positive for the corresponding condition, further comprising administering a therapeutic dose of a therapeutic agent to the subject, wherein the therapeutic agent is selected from the group consisting of obeticholic acid (OCA), tropifexor, elafibranor, saroglitazar, aramchol, semaglutide, tirzepatide, cotadutide, NGM282, MSDC-0602K, resmetirom, cenicriviroc, selonsertib, emricasan, simtuzumab, and GR-MD-02.
[0251] Embodiment 24. The method of any one of embodiments 1 to 18, wherein the corresponding condition is an HCC condition and the diagnostic output identifies the biological sample as positive for the corresponding condition, and further comprising administering a therapeutic dose of a therapeutic agent to the subject, wherein the therapeutic agent is selected from the group consisting of atezolizumab, bevacizumab, sorafenib, lenvatinib, nivolumab, regorafenib, cabozantinib, pemigatinib, ramucirumab, or pembrolizumab.
[0252] Embodiment 25. A method for training a model to diagnose one of a plurality of conditions associated with fatty liver disease (FLD) progression, the method comprising: receiving quantification data for a panel of peptide structures of a plurality of subjects diagnosed with the plurality of conditions associated with FLD progression, the quantification data comprising a plurality of peptide structure profiles of the plurality of subjects, and identifying a corresponding state of the plurality of conditions for each peptide structure profile of the plurality of peptide structure profiles; and training a machine learning model using the quantification data to determine which state of the plurality of conditions a biological sample from the subject corresponds to.
[0253] Embodiment 26. The method of embodiment 25, wherein the machine learning model comprises a logistic regression model.
[0254] Embodiment 27. The method of embodiment 26, wherein the logistic regression model comprises a Lasso regression model.
[0255] Embodiment 28. The method of any one of embodiments 25 to 27, wherein training the machine learning model comprises training the machine learning model using a portion of the quantification data corresponding to a set of peptide structures that is a subset of the panel of peptide structures to determine which of the plurality of states the biological sample from the subject corresponds to.
[0256] Embodiment 29. The method of embodiment 28, further comprising performing differential expression analysis using said quantification data of said plurality of subjects.
[0257] Embodiment 30. The method of embodiment 29, further comprising identifying the set of peptide structures as the subset of the plurality of peptide structures that is relevant to determining to which of the plurality of states the biological sample from the subject corresponds based on at least one of a fold change, a false discovery rate, or a p-value calculated as part of the differential expression analysis.
[0258] Embodiment 31. The method of any one of embodiments 25-30, wherein the plurality of conditions further comprises at least one of a non-alcoholic steatohepatitis (NASH) condition, a hepatocellular carcinoma (HCC) condition, or a healthy condition.
[0259] Embodiment 32. The method of any one of embodiments 25 to 31, wherein the quantification data of the panel of peptide structures for the plurality of subjects diagnosed with the plurality of conditions comprises at least one of abundance, relative abundance, normalized abundance, relative amount, adjusted amount, normalized amount, relative concentration, adjusted concentration, or normalized concentration.
[0260] Embodiment 33. The method of any one of embodiments 25 to 31, wherein the quantification data of the panel of peptide structures for the plurality of subjects diagnosed with the plurality of conditions comprises normalized abundance.
[0261] Embodiment 34. The method of any one of embodiments 25 to 33, wherein the biological sample comprises at least one of blood, serum, or plasma.
[0262] Embodiment 35. The method of any one of embodiments 25 to 34, wherein the quantification data is generated using a liquid chromatography / mass spectrometry (LC / MS) system.
[0263] Embodiment 36. The method of any one of embodiments 25 to 35, wherein the quantification data is generated using multiple reaction monitoring mass spectrometry (MRM-MS).
[0264] Embodiment 37. A method for detecting the presence of one of a plurality of conditions associated with fatty liver disease (FLD) progression in a biological sample, the method comprising: receiving peptide structure data corresponding to a set of glycoproteins in the biological sample obtained from a subject; analyzing the peptide structure data using a supervised machine learning model to generate a disease indicator based on at least three peptide structures selected from the group of peptide structures identified in Table 1; and detecting the presence of the corresponding condition among the plurality of conditions associated with FLD progression in response to determining that the disease indicator falls within a selected range associated with the corresponding condition.
[0265] Embodiment 38. The method of embodiment 37, wherein a peptide structure of the at least three peptide structures comprises a glycopeptide structure defined by a peptide sequence identified in Table 1 and a glycan structure linked to the peptide sequence at a linkage site of the peptide sequence, wherein the peptide sequence is one of SEQ ID NOs: 23, 24, 25, 29, 30, 31, 33, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 52, or 52 as defined in Table 1.
[0266] Embodiment 39. The method of embodiment 37 or embodiment 38, wherein the supervised machine learning model comprises a logistic regression model.
[0267] Embodiment 40. The method of any one of embodiments 37 to 39, wherein the supervised machine learning model comprises a penalized multivariate logistic regression model.
[0268] Embodiment 41. The method of any one of embodiments 37 to 40, wherein the peptide structure data comprises quantification data.
[0269] Embodiment 42. The method of embodiment 41, wherein the quantification data for a peptide structure in the group of peptide structures comprises at least one of abundance, relative abundance, normalized abundance, relative amount, adjusted amount, normalized amount, relative concentration, adjusted concentration, or normalized concentration.
[0270] Embodiment 43. The method of embodiment 41, wherein the quantification data comprises normalized abundance.
[0271] Embodiment 44. The method of any one of embodiments 37 to 43, wherein the disease indicator is a probability score.
[0272] Embodiment 45. The method of any one of embodiments 37 to 44, further comprising generating a report including a diagnosis based on the corresponding condition detected for the subject.
[0273] Embodiment 46. The method of any one of embodiments 37-45, wherein the plurality of conditions comprises at least two selected from the group consisting of a non-alcoholic steatohepatitis (NASH) condition, a hepatocellular carcinoma (HCC) condition, and a non-NASH / HCC condition.
[0274] Embodiment 47. The method of any one of embodiments 37 to 46, wherein analyzing the peptide structure data includes calculating the disease indicator using a weighting factor associated with each peptide structure of the at least three peptide structures, wherein the weighting factor of a corresponding peptide structure of the at least three peptide structures indicates the relative significance of the corresponding peptide structure for the disease indicator.
[0275] Embodiment 48. The method of any one of embodiments 37 to 46, wherein analyzing the peptide structures comprises calculating a peptide structure profile of the biological sample identifying a weighted value for each of the at least three peptide structures, wherein the weighted value for each of the at least three peptide structures is a product of a quantification metric for the peptide structure identified from the peptide structure data and a weighting coefficient for the peptide structure; and calculating the disease indicator using the peptide structure profile.
[0276] Embodiment 49. The method of any one of embodiments 37 to 48, wherein the corresponding state is a non-NASH / HCC state and the selected range associated with the non-NASH / HCC state is 0.00 to 0.05.
[0277] Embodiment 50. The method of any one of embodiments 37 to 48, wherein the corresponding condition is a non-alcoholic steatohepatitis (NASH) condition and the selected range associated with the NASH condition is 0.05 to 0.4.
[0278] Embodiment 51. The method of any one of embodiments 37 to 48, wherein the corresponding condition is a hepatocellular carcinoma (HCC) condition and the selected range associated with the HCC condition is 0.4 to 1.0.
[0279] Embodiment 52. The method of any one of embodiments 37 to 51, further comprising creating a sample from the biological sample and preparing the sample using reduction, alkylation, and enzymatic digestion to form a prepared sample comprising a set of peptide structures.
[0280] Embodiment 53. The method of embodiment 52, further comprising generating said peptide structural data from said prepared sample using multiple reaction monitoring mass spectrometry (MRM-MS).
[0281] Embodiment 54. The method of any one of embodiments 37 to 52, wherein the peptide structural data is generated using a liquid chromatography / mass spectrometry (LC / MS) system.
[0282] Embodiment 55. The method of any one of embodiments 37 to 52, wherein the peptide structural data is generated using multiple reaction monitoring mass spectrometry (MRM-MS).
[0283] Embodiment 56. The method of any one of embodiments 37 to 55, wherein the biological sample comprises at least one of blood, serum, or plasma.
[0284] Embodiment 57. The method of any one of embodiments 37 to 56, further comprising generating a treatment output based on the disease indicator.
[0285] Embodiment 58. The method of embodiment 57, wherein the treatment output includes at least one of an identification of a treatment for treating the subject, a design for the treatment, a manufacturing plan for the treatment, or a treatment plan for administering the treatment.
[0286] Embodiment 59. The method of embodiment 58, wherein the corresponding condition is a NASH condition and the treatment comprises at least one of obeticholic acid (OCA), tropifexor, elafibranor, saroglitazar, aramchol, semaglutide, tirzepatide, cotadutide, NGM282, MSDC-0602K, resmetirom, cenicriviroc, selonsertib, emricasan, simtuzumab, and GR-MD-02.
[0287] Embodiment 60. The method of embodiment 58, wherein the corresponding condition is an HCC condition and the treatment comprises at least one of atezolizumab, bevacizumab, sorafenib, lenvatinib, nivolumab, regorafenib, cabozantinib, pemigatinib, ramucirumab, or pembrolizumab.
[0288] Embodiment 61. The method of any one of embodiments 37 to 57, wherein the corresponding condition is a NASH condition, and further comprising administering to the subject a therapeutic dose of a therapeutic agent, wherein the therapeutic agent is selected from the group consisting of obeticholic acid (OCA), tropifexor, elafibranor, saroglitazar, aramchol, semaglutide, tirzepatide, cotadutide, NGM282, MSDC-0602K, resmetirom, cenicriviroc, selonsertib, emricasan, simtuzumab, and GR-MD-02.
[0289] Embodiment 62. The method of any one of embodiments 37 to 57, wherein the corresponding condition is an HCC condition, and further comprising administering to the subject a therapeutic dose of a therapeutic agent, wherein the therapeutic agent is selected from the group consisting of atezolizumab, bevacizumab, sorafenib, lenvatinib, nivolumab, regorafenib, cabozantinib, pemigatinib, ramucirumab, or pembrolizumab.
[0290] Embodiment 63. A method for classifying a biological sample as corresponding to one of a plurality of states associated with fatty liver disease (FLD) progression, comprising: training a supervised machine learning model using training data, the training data including a plurality of peptide structure profiles of a plurality of training subjects, and identifying one of the plurality of states for each peptide structure profile of the plurality of peptide structure profiles; receiving peptide structure data corresponding to a set of glycoproteins in the biological sample obtained from a subject; and comparing quantification data identified from the peptide structure data for the set of peptide structures with the trained supervised machine learning model. 10. The method of claim 1, further comprising: inputting the set of peptide structures into a supervised machine learning model, wherein the set of peptide structures includes at least one peptide structure identified in Table 1; analyzing the quantification data using the supervised machine learning model to generate a score; determining that the score falls within a selected range associated with a corresponding state among the plurality of states associated with the FLD progression; and generating a diagnostic output indicating that the biological sample demonstrates the corresponding state, wherein the plurality of states include a non-alcoholic steatohepatitis (NASH) state, a hepatocellular carcinoma (HCC) state, and a non-NASH / HCC state.
[0291] Embodiment 64. A method of treating a non-alcoholic steatohepatitis (NASH) disorder in a patient to at least one of reduce, stall, or reverse progression of the NASH disorder to hepatocellular carcinoma, comprising: receiving a biological sample from the patient; determining the amount of each peptide structure identified in Table 1 in the biological sample using a multiple reaction monitoring mass spectrometry (MRM-MS) system; analyzing the amount of each peptide structure using a machine learning model to generate a disease indicator; generating a diagnostic output based on the disease indicator that classifies the biological sample as evidence that the patient has the NASH disorder; and administering obeticholic acid (OCA) or a derivative thereof to the patient, wherein the administering comprises at least one of intravenous and oral administration in the range of 10 to 25 mg daily.
[0292] Embodiment 65. The method of embodiment 64, further comprising preparing the biological sample to form a prepared sample comprising a set of peptide structures, and inputting the prepared sample into the MRM-MS system using a liquid chromatography system.
[0293] Embodiment 66. A method of treating a hepatocellular carcinoma (HCC) disorder in a patient, comprising receiving the biological sample from the patient; determining an amount of each peptide structure identified in Table 1 in the biological sample using a multiple reaction monitoring mass spectrometry (MRM-MS) system; analyzing the amount of each peptide structure using a machine learning model to generate a disease indicator; and generating a diagnostic output based on the disease indicator that classifies the biological sample as evidence that the patient has the HCC disorder. sorafenib or a derivative thereof to said patient, said administration comprising at least one of intravenous and oral administration in the range of 775-825 mg daily; lenvatinib or a derivative thereof to a patient, said administration comprising at least one of intravenous and oral administration in the range of 7.5 to 8.5 mg / day when said patient weighs less than 60 kg and 11.5 to 12.5 mg / day when said patient weighs more than 60 kg; Nivolumab or a derivative thereof to the patient, said administration including at least one of intravenous and oral administration in the range of 0.75 to 1.25 mg / kg; regorafenib or a derivative thereof to the patient), wherein the administration includes oral administration in the range of 150 to 170 mg / day; Cabozantinib or a derivative thereof to the patient, the administration comprising at least one of intravenous and oral administration in the range of 50-70 mg / day; administering to the patient at least one of ramucirumab or a derivative thereof, wherein the administration comprises at least one of intravenous and oral administration in the range of 8-12 mg / kg.
[0294] Embodiment 67. The method of embodiment 66, further comprising preparing the biological sample to form a prepared sample comprising a set of peptide structures, and inputting the prepared sample into the MRM-MS system for analysis using a liquid chromatography system.
[0295] Embodiment 68. A method for designing a treatment for a subject diagnosed with a disease state associated with fatty liver disease (FLD) progression, comprising designing a therapeutic agent for treating the subject in response to determining that a biological sample obtained from the subject evidences the disease state using some or all of the methods described in any one of embodiments 1 to 63.
[0296] Embodiment 69. A method for planning treatment of a subject diagnosed with a disease state associated with fatty liver disease (FLD) progression, comprising generating a treatment plan for treating the subject in response to determining that a biological sample obtained from the subject evidences the disease state using some or all of the methods described in any one of embodiments 1 to 63.
[0297] Embodiment 70. A method of treating a subject diagnosed with a disease state associated with fatty liver disease (FLD) progression, comprising administering to the subject a therapeutic agent to treat the subject based on determining that a biological sample obtained from the subject evidences the disease state using some or all of the methods described in any one of embodiments 1 to 63.
[0298] Embodiment 71. A method of treating a subject diagnosed with a disease state associated with fatty liver disease (FLD) progression, comprising selecting a therapeutic agent for treating the subject based on determining that the subject will respond to the therapeutic agent using the method of any one of embodiments 1 to 63.
[0299] Embodiment 72. A method for analyzing a set of peptide structures in a sample from a patient, said method comprising: (a) obtaining said sample from said patient; (b) preparing said sample to form a prepared sample comprising said set of peptide structures; (c) inputting the prepared sample into a mass spectrometry system using a liquid chromatography system; (d) detecting a set of product ions associated with each peptide structure of the set of peptide structures using the mass spectrometry system; the set of peptide structures comprises at least one peptide structure selected from peptide structures PS-1 to PS-53 identified in Table 4; The set of peptide structures is a peptide structure comprising: (i) precursor ions having a mass-to-charge (m / z) ratio within ±1.5 of the m / z ratio listed for the precursor ion in Table 4 as corresponding to said peptide structure; (ii) a product ion having an m / z ratio that is within ±1.0 of the m / z ratio listed for the first product ion in Table 4 as corresponding to said peptide structure; (e) generating quantification data for the set of product ions using the mass spectrometry system.
[0300] Embodiment 73. The method of embodiment 72, wherein the mass-to-charge (m / z) ratio of the precursor ion is within ±1.0 of the m / z ratio listed for the precursor ion in Table 4 as corresponding to the peptide structure.
[0301] Embodiment 74. The method of embodiment 72, wherein the mass-to-charge (m / z) ratio of the precursor ion is within ±0.5 of the m / z ratio listed for the precursor ion in Table 4 as corresponding to the peptide structure.
[0302] Embodiment 75. The method of embodiment 72, wherein the mass-to-charge (m / z) ratio of the product ions is within ±0.8 of the m / z ratio listed for precursor ions in Table 4 as corresponding to the peptide structure.
[0303] Embodiment 76. The method of embodiment 72, wherein the mass-to-charge (m / z) ratio of the product ions is within ±0.5 of the m / z ratio listed for precursor ions in Table 4 as corresponding to the peptide structure.
[0304] Embodiment 77. The method of any one of embodiments 72 to 76, further comprising generating a diagnostic output using the quantification data and a machine learning model trained to detect the presence of a disease state associated with fatty liver disease (FLD) progression.
[0305] Embodiment 78. The method of any one of embodiments 72 to 77, wherein the mass spectrometry system comprises a reaction monitoring mass spectrometry system that detects the set of product ions and generates the quantification data using at least one of multiple reaction monitoring mass spectrometry (MRM-MS) or selected reaction monitoring mass spectrometry (SRM-MS).
[0306] Embodiment 79. The method of any one of embodiments 72 to 78, wherein the liquid chromatography system is a high-performance liquid chromatography system.
[0307] Embodiment 80. The method of any one of embodiments 72 to 79, wherein the sample comprises a plasma sample.
[0308] Embodiment 81. The method of any one of embodiments 72 to 80, wherein the sample comprises a serum sample.
[0309] Embodiment 82. The method of any one of embodiments 72 to 81, wherein the sample comprises a blood sample.
[0310] Embodiment 83. The method of any one of embodiments 72 to 82, wherein preparing the sample comprises at least one of: denaturing one or more proteins in the sample to form one or more denatured proteins; reducing the one or more denatured proteins in the sample to form one or more reduced proteins; alkylating one or more proteins in the sample using an alkylating agent to prevent reformation of disulfide bonds in the one or more reduced proteins to form one or more alkylated proteins; or digesting the one or more alkylated proteins in the sample using a proteolytic catalyst to form the prepared sample comprising the set of peptide structures.
[0311] Embodiment 84. A composition comprising at least one of the peptide structures PS-1 to PS-53 identified in Table 1.
[0312] Embodiment 85. A composition comprising a peptide structure or product ion, wherein the peptide structure or product ion comprises an amino acid sequence having at least 90% sequence identity to any one of SEQ ID NOs: 23-52, which correspond to peptide structures PS-1 to PS-53 in Table 1, and the product ion is selected as one from the group consisting of the product ions identified in Table 4, including product ions that fall within the specified m / z ranges.
[0313] Embodiment 86. A composition comprising a glycopeptide structure selected as one from the group consisting of peptide structures PS-1 to PS-53 identified in Table 4, wherein the glycopeptide structure comprises an amino acid peptide sequence identified in Table 5 as corresponding to said glycopeptide structure, and a glycan structure identified in Table 1 as corresponding to the glycopeptide structure, wherein said glycan structure is linked to residues of the amino acid peptide sequence at corresponding positions identified in Table 1, and wherein the glycan structure has a glycan composition.
[0314] Embodiment 87. The composition of embodiment 86, wherein the glycan composition is specified in Table 7.
[0315] Embodiment 88. The composition of embodiment 86 or embodiment 87, wherein the glycopeptide structure has a precursor ion whose charge is identified in Table 4 as corresponding to the glycopeptide structure.
[0316] Embodiment 89. The composition of any one of embodiments 86-88, wherein the glycopeptide structure has a precursor ion having an m / z ratio within ±1.5 of the m / z ratio listed for the precursor ion in Table 4 as corresponding to the glycopeptide structure.
[0317] Embodiment 90. The composition of any one of embodiments 86-88, wherein the glycopeptide structure has a precursor ion having an m / z ratio within ±1.0 of the m / z ratio listed for the precursor ion in Table 4 as corresponding to the glycopeptide structure.
[0318] Embodiment 91. The composition of any one of embodiments 86-88, wherein the glycopeptide structure has a precursor ion having an m / z ratio within ±0.5 of the m / z ratio listed for the precursor ion in Table 4 as corresponding to the glycopeptide structure.
[0319] Embodiment 92. The composition of any one of embodiments 86-91, wherein the glycopeptide structure has a product ion having an m / z ratio within ±1.0 of the m / z ratio listed for the first product ion in Table 4 as corresponding to the glycopeptide structure.
[0320] Embodiment 93. The composition of any one of embodiments 86-91, wherein the glycopeptide structure has a product ion having an m / z ratio within ±0.8 of the m / z ratio listed for the first product ion in Table 4 as corresponding to the glycopeptide structure.
[0321] Embodiment 94. The composition of any one of embodiments 86-91, wherein the glycopeptide structure has a product ion having an m / z ratio within ±0.5 of the m / z ratio listed for the first product ion in Table 4 as corresponding to the glycopeptide structure.
[0322] Embodiment 95. The composition of any one of embodiments 86-94, wherein the glycopeptide structure has a monoisotopic mass identified in Table 1 as corresponding to said glycopeptide structure.
[0323] Embodiment 96. A composition comprising a peptide structure selected from a plurality of peptide structures identified in Table 1, wherein the peptide structure has a monoisotopic mass identified in Table 1 as corresponding to the peptide structure, and the peptide structure comprises an amino acid sequence of SEQ ID NO: 23-52 identified in Table 1 as corresponding to the peptide structure.
[0324] Embodiment 97. The composition of embodiment 96, wherein the peptide structure has a precursor ion whose charge is identified in Table 4 as corresponding to the peptide structure.
[0325] Embodiment 98. The composition of embodiment 96 or embodiment 97, wherein the peptide structure has a precursor ion having an m / z ratio within ±1.5 of the m / z ratio listed for the precursor ion in Table 4 as corresponding to the peptide structure.
[0326] Embodiment 99. The composition of embodiment 96 or embodiment 97, wherein the peptide structure has a precursor ion having an m / z ratio within ±1.0 of the m / z ratio listed for the precursor ion in Table 4 as corresponding to the peptide structure.
[0327] Embodiment 100. The composition of embodiment 96 or embodiment 97, wherein the peptide structure has a precursor ion having an m / z ratio within ±0.5 of the m / z ratio listed for the precursor ion in Table 4 as corresponding to the peptide structure.
[0328] Embodiment 101. The composition of any one of embodiments 96-100, wherein the peptide structure has a product ion having an m / z ratio within ±1.0 of the m / z ratio listed for the first product ion in Table 4 as corresponding to the peptide structure.
[0329] Embodiment 102. The composition of any one of embodiments 96 to 100, wherein the peptide structure has a product ion having an m / z ratio within ±0.8 of the m / z ratio listed for the first product ion in Table 4 as corresponding to the peptide structure.
[0330] Embodiment 103. The composition of any one of embodiments 96 to 100, wherein the peptide structure has a product ion having an m / z ratio within ±0.5 of the m / z ratio listed for the first product ion in Table 4 as corresponding to the peptide structure.
[0331] Embodiment 104. A kit for carrying out part or all of the method according to any one of embodiments 1 to 83, comprising at least one agent for quantifying at least one peptide structure identified in Table 1.
[0332] Embodiment 105. A kit for carrying out part or all of the method according to any one of embodiments 1 to 83, comprising at least one of a glycopeptide standard, a buffer, or a set of peptide sequences, wherein a peptide sequence in the set of peptide sequences is identified by a corresponding one of SEQ ID NOs: 23 to 52 as defined in Table 1.
[0333] Embodiment 106. A system comprising one or more data processors and a non-transitory computer-readable storage medium comprising instructions that, when executed on the one or more data processors, cause the one or more data processors to perform some or all of the functions described in any one of embodiments 1 to 83.
[0334] Embodiment 107. A computer program product tangibly embodied in a non-transitory machine-readable storage medium, the computer program product comprising instructions configured to cause one or more data processors to perform some or all of the functions described in any one of embodiments 1 to 83.
[0335] Furthermore, an embodiment may include some or all of any one or more of the above-described embodiments 1 to 107. For example, an embodiment may include one or more features from any one or more of embodiments 1 to 107, used in any combination.
[0336] XI. Additional Considerations Any headings and / or subheadings between sections and subsections of this document are included solely for the purpose of improving readability and do not imply that features cannot be combined across sections and subsections, and therefore, the sections and subsections do not describe separate embodiments.
[0337] While the present teachings will be described in conjunction with various embodiments, it is not intended that the present teachings be limited to such embodiments. On the contrary, the present teachings encompass various alternatives, modifications, and equivalents, as will be appreciated by those skilled in the art. The present description provides preferred exemplary embodiments and is not intended to limit the scope, applicability, or configuration of the present disclosure. Rather, the present description of preferred exemplary embodiments will provide those skilled in the art with an effective description for practicing various embodiments.
[0338] It is understood that various changes may be made in the function and arrangement of elements without departing from the spirit and scope as set forth in the appended claims. Therefore, such modifications and variations are deemed to be within the scope as set forth in the appended claims. Furthermore, the terms and expressions employed are used as words of description and not of limitation, and there is no intention in the use of such terms and expressions to exclude any equivalents of the features shown and described or portions thereof, recognizing that various modifications are possible within the scope of the invention as claimed.
[0339] In describing various embodiments, the specification may present a method and / or process as a particular sequence of steps. However, to the extent that the method or process does not rely on the particular order of steps set forth herein, the method or process should not be limited to the particular order of steps described, and one of ordinary skill in the art will readily recognize that the sequence can be varied and still remain within the spirit and scope of the various embodiments.
[0340] Some embodiments of the present disclosure include a system including one or more data processors. In some embodiments, the system includes a non-transitory computer-readable storage medium including instructions that, when executed on the one or more data processors, cause the one or more data processors to perform some or all of one or more methods and / or some or all of one or more processes disclosed herein. Some embodiments of the present disclosure include a computer program product tangibly embodied in a non-transitory machine-readable storage medium including instructions configured to cause one or more data processors to perform some or all of one or more methods and / or some or all of one or more processes disclosed herein.
[0341] Specific details are provided herein to provide an understanding of the embodiments. However, it will be understood that the embodiments may be practiced without these specific details. For example, circuits, systems, networks, processes, and other components may be shown as components in block diagram form in order to avoid obscuring the embodiments in unnecessary detail. In other instances, well-known circuits, processes, algorithms, structures, and techniques may be shown without unnecessary detail in order to avoid obscuring the embodiments.
Claims
1. 1. A method for classifying a biological sample with respect to a plurality of conditions associated with fatty liver disease (FLD) progression, comprising: receiving peptide structure data corresponding to a set of glycoproteins in the biological sample obtained from a subject; inputting quantification data identified from the peptide structure data for a set of peptide structures into a machine learning model, the set of peptide structures including at least one peptide structure identified from a plurality of peptide structures shown in Table 1 below; analyzing the quantification data using the machine learning model to generate a disease indicator; generating a diagnostic output based on the disease indicator that classifies the biological sample as evidencing a corresponding condition among the plurality of conditions associated with the FLD progression. [Table 1]
2. wherein the disease indicator comprises a score, and generating the diagnostic output comprises: determining that the score falls within a selected range associated with the corresponding condition of the plurality of conditions; and determining that the biological sample evidences the corresponding condition in response to determining that the score falls within the selected range associated with the corresponding condition.
3. 10. The method of claim 1, wherein the plurality of conditions comprises a non-alcoholic steatohepatitis (NASH) condition and a hepatocellular carcinoma (HCC) condition.
4. 2. The method of claim 1, wherein the plurality of conditions comprises non-NASH / HCC conditions, including at least one of a healthy condition, a condition without liver disease, or a benign liver mass condition.
5. the at least one peptide structure comprises a glycopeptide structure defined by a peptide sequence and a glycan structure linked to the peptide sequence at a linkage site of the peptide sequence; the peptide sequence is one of SEQ ID NOs: 23, 24, 25, 29, 30, 31, 33, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, or 52 as defined in Table 1 above; The glycan structure corresponds to the GL number shown in Table 2 below, 2. The method of claim 1, wherein the linkage site in the peptide sequence to the glycan structure is shown in Table 2 below as the position of the linkage site within the peptide sequence. [Table 2]
6. 2. The method of claim 1, wherein the quantification data for a peptide structure in the set of peptide structures comprises at least one of abundance, relative abundance, normalized abundance, relative amount, adjusted amount, normalized amount, relative concentration, adjusted concentration, or normalized concentration.
7. The method of claim 1 , wherein the peptide structural data is generated using multiple reaction monitoring mass spectrometry (MRM-MS).
8. 1. A method for detecting the presence of one of a number of conditions associated with fatty liver disease (FLD) progression in a biological sample, comprising: receiving peptide structure data corresponding to a set of glycoproteins in the biological sample obtained from a subject; analyzing the peptide structure data using a supervised machine learning model to generate disease indicators based on at least three peptide structures selected from the group of peptide structures identified in Table 3 below; and in response to determining that the disease indicator falls within a selected range associated with the corresponding condition, detecting the presence of a corresponding condition among the plurality of conditions associated with the FLD progression. [Table 3]
9. a peptide structure of the at least three peptide structures comprises a glycopeptide structure defined by a peptide sequence and a glycan structure linked to the peptide sequence at a linkage site of the peptide sequence, the peptide sequence being one of SEQ ID NOs: 23, 24, 25, 29, 30, 31, 33, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 52, or 52 as defined in Table 3 above; The glycan structure corresponds to the GL number shown in Table 4 below, 9. The method of claim 8, wherein the linkage site in the peptide sequence to the glycan structure is shown in Table 4 below as the position of the linkage site within the peptide sequence. [Table 4]
10. The method of claim 8 , wherein the supervised machine learning model comprises a logistic regression model.
11. The method of claim 8 , wherein the supervised machine learning model comprises a penalized multivariate logistic regression model.
12. 9. The method of claim 8, wherein the peptide structure data comprises quantification data, and the quantification data for a peptide structure in the group of peptide structures comprises at least one of abundance, relative abundance, normalized abundance, relative amount, adjusted amount, normalized amount, relative concentration, adjusted concentration, or normalized concentration.
13. The method of claim 8 , wherein the disease indicator is a probability score.
14. The method of claim 8 , further comprising generating a report including a diagnosis based on the corresponding condition detected for the subject.
15. 9. The method of claim 8, wherein the plurality of conditions comprises at least two selected from the group consisting of a non-alcoholic steatohepatitis (NASH) condition, a hepatocellular carcinoma (HCC) condition, and a non-NASH / HCC condition.
16. The method of claim 1, wherein the at least one peptide structure comprises a glycopeptide structure defined by a peptide sequence defined in Table 1 and a glycan structure linked to the peptide sequence at a linkage site of the peptide sequence; The glycan structure corresponds to the GL number shown in Table 5 below, 2. The method of claim 1, wherein the linkage site in the peptide sequence to the glycan structure is shown in Table 5 below as the position of the linkage site within the peptide sequence. [Table 5] 17. The method of claim 17, wherein a peptide structure of the at least three peptide structures comprises a glycopeptide structure defined by a peptide sequence defined in Table 3 and a glycan structure linked to the peptide sequence at a linkage site of the peptide sequence; The glycan structure corresponds to the GL number shown in Table 6 below, 9. The method of claim 8, wherein the linkage site in the peptide sequence to the glycan structure is shown in Table 6 below as the position of the linkage site within the peptide sequence. [Table 6]
18. The method described in claim 5, 9, 16 or 17, wherein the glycan structure comprises the glycan composition shown in Table 7 below. [Table 7]
