Pet relative search system and method

Abstract

A computer-implemented method for inferring companion pet kinship using companion pet DNA information, the method comprising: receiving companion pet DNA information including at least one DNA sequence; transmitting the at least one DNA sequence to a database, wherein the database receiving the at least one DNA sequence searches for and outputs at least one local DNA fragment pattern including at least one matching chromosome segment corresponding to the at least one matched companion pet; identifying at least one inferred kinship between the companion pet and the at least one matched companion pet by analyzing the at least one local DNA fragment pattern; and outputting at least one inferred kinship based on the analyzing step.

Description

[Technical field]

[0001] Various embodiments of the present disclosure relate generally to machine learning based techniques for inferring kinship among companion pets, and in some embodiments, the present disclosure relates to systems and methods for training machine learning based models to infer kinship among companion pets. [Background technology]

[0002] Identifying the familial relationship of companion pets is a very difficult and erroneous process. For example, a conventional technique includes a step of determining whether two companion pets are related by comparing similar DNA fragments between the two companion pets. In the conventional technique, if two companion pets share similar DNA, it may be concluded that the two companion pets are related by blood. However, the conventional technique cannot distinguish between DNA fragments indicating inbreeding and DNA fragments indicating kinship. Therefore, the conventional technique determines that two companion pets are related by blood when they are actually only inbreeding. Furthermore, the conventional technique does not have a function to provide a user with kinship (relationship) labels (e.g., labels such as mother, sister, half-sister, nephew, etc.) for two or more companion pets.

[0003] The present disclosure is intended to address the above-mentioned problems. Note that the description of the background art in this section is intended to provide a brief overview of the background of the present disclosure. Unless otherwise specified in this specification, the contents described in this section are not prior art to the claims of the present application. Therefore, the descriptions in this section are not acknowledged as prior art or suggest prior art. Summary of the Invention

[0004] According to certain aspects of the present disclosure, a method and system for inferring kinship of companion pets is disclosed.

[0005] In one aspect, an exemplary embodiment of a computer-implemented method for inferring companion pet kinship using companion pet DNA information includes receiving companion pet DNA information including at least one DNA sequence; transmitting the at least one DNA sequence to a database, wherein the database receiving the at least one DNA sequence searches for and outputs at least one local DNA fragment pattern including at least one matching chromosome segment corresponding to the at least one matched companion pet; identifying at least one inferred kinship relationship between the companion pet and the at least one matched companion pet by analyzing the at least one local DNA fragment pattern; and outputting at least one inferred kinship relationship based on the analyzing step.

[0006] In one aspect, an exemplary embodiment of a computer system for inferring companion pet kinship using companion pet DNA information comprises at least one memory storing instructions and at least one processor configured to perform operations by executing the instructions, including receiving companion pet DNA information including at least one DNA sequence, transmitting the at least one DNA sequence to a database, where the database receiving the at least one DNA sequence retrieves and outputs at least one local DNA fragment pattern including at least one matching chromosome segment corresponding to the matched at least one companion pet, identifying at least one inferred kinship between the companion pet and the matched at least one companion pet by analyzing the at least one local DNA fragment pattern, and outputting at least one inferred kinship based on the analyzing step.

[0007] In one aspect, an exemplary embodiment of a non-transitory computer readable medium contains instructions that, when executed by a processor, cause the processor to perform operations for inferring kinship of companion pets using DNA information of companion pets, the operations including receiving DNA information of companion pets including at least one DNA sequence, transmitting the at least one DNA sequence to a database, where the database receiving the at least one DNA sequence retrieves and outputs at least one local DNA fragment pattern including at least one matching chromosome segment corresponding to the at least one matched companion pet, identifying at least one inferred kinship between the companion pet and the at least one matched companion pet by analyzing the at least one local DNA fragment pattern, and outputting at least one inferred kinship based on the analyzing step.

[0008] In one aspect, an exemplary embodiment of a method for training a machine learning system to infer kinship relationships of companion pets using DNA information of companion pets can include receiving DNA information of companion pets (including at least one DNA sequence, at least one matched companion pet, and at least one relationship label corresponding to a kinship relationship between the companion pet and the at least one matched companion pet) and training a machine learning model to infer at least one kinship relationship from the at least one DNA sequence in response to the receiving step. The training step can also include extracting at least one local DNA fragment pattern corresponding to the at least one matched companion pet from a database for each of the at least one DNA sequence, identifying at least one inferred kinship relationship between the companion pet and the at least one matched companion pet by analyzing the at least one local DNA fragment pattern, and identifying the at least one inferred kinship relationship and at least one corresponding confidence level based on the analysis.

[0009] In a further aspect, an exemplary embodiment of a computer system for training a machine learning system to infer kinship relationships of companion pets using DNA information of companion pets may comprise at least one memory storing instructions and at least one processor configured to perform operations by executing the instructions, the operations including receiving DNA information of companion pets (including at least one DNA sequence, at least one matched companion pet, and at least one relationship label corresponding to a kinship relationship between the companion pet and the at least one matched companion pet) and training a machine learning model to infer at least one kinship relationship from the at least one DNA sequence in response to the receiving step. The training step may also include extracting, for each of the at least one DNA sequence, at least one local DNA fragment pattern from the database corresponding to the at least one matched companion pet; identifying at least one inferred kinship relationship between the companion pet and the at least one matched companion pet by analyzing the at least one local DNA fragment pattern; and identifying the at least one inferred kinship relationship and at least one corresponding confidence level based on the analysis.

[0010] In a further aspect, a non-transitory computer readable medium can contain instructions that, when executed by a processor, cause the processor to perform operations for training a machine learning system to infer kinship relationships of companion pets using DNA information of companion pets, the operations can include receiving DNA information of companion pets (including at least one DNA sequence, at least one matched companion pet, and at least one relationship label corresponding to a kinship relationship between the companion pet and the at least one matched companion pet), and training a machine learning model to infer at least one kinship relationship from the at least one DNA sequence in response to the receiving step. The training step may also include extracting, for each of the at least one DNA sequence, at least one local DNA fragment pattern from the database corresponding to the at least one matched companion pet; identifying at least one inferred kinship relationship between the companion pet and the at least one matched companion pet by analyzing the at least one local DNA fragment pattern; and identifying the at least one inferred kinship relationship and at least one corresponding confidence level based on the analysis.

[0011] It is to be understood that the foregoing general description and the following detailed description are merely illustrative of embodiments of the present disclosure as set forth in the claims, and are not intended to limit the disclosure of the embodiments of the present disclosure as set forth in the claims. [Brief description of the drawings]

[0012] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate various exemplary embodiments and, together with the following description, serve to explain the principles of embodiments of the present disclosure. [Figure 1A] FIG. 1 illustrates an example environment of a platform for inferring companion pet kinship, according to one or more embodiments. [Figure 1B] FIG. 1 illustrates an example environment of a platform for inferring companion pet kinship, according to one or more embodiments. [Figure 1C] FIG. 1 illustrates an example environment of a platform for inferring companion pet kinship, according to one or more embodiments. [Figure 1D] FIG. 1 illustrates an example environment of a platform for inferring companion pet kinship, according to one or more embodiments. [Figure 1E] FIG. 1 illustrates an example environment of a platform for inferring companion pet kinship, according to one or more embodiments. [Figure 1F] FIG. 1 illustrates an example environment of a platform for inferring companion pet kinship, according to one or more embodiments. [Figure 1G] FIG. 1 illustrates an example environment of a platform for inferring companion pet kinship, according to one or more embodiments. [Figure 1H] FIG. 1 illustrates an example environment of a platform for inferring companion pet kinship, according to one or more embodiments. [Figure 1I] FIG. 1 illustrates an example environment of a platform for inferring companion pet kinship, according to one or more embodiments. [Figure 1J] FIG. 1 illustrates an example environment of a platform for inferring companion pet kinship, according to one or more embodiments. [Diagram 2] 1 is a flowchart illustrating an exemplary method for training a machine learning model to infer kinship among companion pets, according to one or more embodiments. [Diagram 3] FIG. 1 is a flow chart illustrating an exemplary method for training a machine learning model to identify at least one inferred kinship relationship by analyzing at least one local DNA fragment pattern, according to one or more embodiments. [Figure 4]1 is a flowchart further illustrating an exemplary method for training a machine learning model to infer at least one shared trait between a companion pet and a matched companion pet, according to one or more embodiments. [Diagram 5] FIG. 1 is a flowchart further illustrating an exemplary method for training a machine learning model to output a list of DNA-matched companion pet samples, according to one or more embodiments. [Figure 6] FIG. 1 illustrates an exemplary process for inferring companion pet kinship using DNA information of companion pets, according to one or more embodiments. [Figure 7] FIG. 1 illustrates an example environment in which the techniques presented herein can be utilized, in accordance with one or more embodiments. [Figure 8] FIG. 1 illustrates an example of a computing device capable of executing the techniques presented herein, in accordance with one or more embodiments. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] According to certain aspects of the present disclosure, a method and system for inferring kinship of companion pets is disclosed. Conventional techniques do not take into account inbreeding of companion pets and are therefore believed to be inadequate for this purpose. Thus, there is a need for improved techniques for inferring kinship of companion pets.

[0014] Genetic data such as DNA fragments have been used to find companion pets that are closely related to each other. For example, such a process involves analyzing long DNA fragments in the genomes of at least two companion pets to determine which DNA fragments are identical. These long identical DNA fragments can be called "identical-by-descent" (IBD) fragments. However, not all identical DNA fragments can be called identical-by-descent fragments. For example, many purebred dogs that are the product of inbreeding have long stretches of identical DNA among their lineage. This is because breeds are created by inbreeding a small number of individuals to express certain characteristics. Furthermore, a process that cannot identify whether a DNA fragment is an identical fragment or the result of inbreeding cannot identify the kinship relationship between these two companion pets. Furthermore, from a genetic perspective, there may be cases where the amount of identical DNA is the same even if the relationship is different, such as between great-grandparents and great-grandchildren, or between uncles, aunts, nieces, and nephews. Therefore, there is a need to identify whether a DNA fragment is a homoeologous fragment or the result of inbreeding, and to identify kinship relationships based on homoeologous DNA.

[0015] As described in more detail below, in various embodiments, systems and methods are described for inferring kinship of companion pets using machine learning models. A machine learning model can be trained (e.g., trained by supervised or semi-supervised learning) to learn associations between DNA information of companion pets (e.g., breed, at least one matched companion pet, at least one relationship label) such that the trained machine learning model can be utilized to infer kinship of companion pets.

[0016] Various aspects of machine learning techniques that can be adapted to infer kinship of companion pets are presented herein. As described in detail below, machine learning techniques adapted to infer kinship of companion pets can include one or more aspects according to the present disclosure, such as, for example, a specific selection method of training data, a specific training process of a machine learning model, the operation of a specific device suitable for use in combination with a trained machine learning model, the operation of a machine learning model in conjunction with specific data, a method of modifying such specific data by a machine learning model, other aspects that would be apparent to a person skilled in the art based on the present disclosure, or any combination thereof.

[0017] Relative search platform 1A-1J illustrate an example environment of a platform for inferring companion pet kinship, according to one or more embodiments.

[0018] FIG. 1A illustrates a “Highlights” page in an exemplary environment of a platform for inferring kinship of companion pets, according to one or more embodiments. The platform may display the name (e.g., Coco), an image, or both of the companion pet, as well as the location (e.g., Portland, Oregon), a social media account (e.g., @cocothegreat), or both of the companion pet. Additionally, the platform may further display a “Like” button (e.g., a heart button), one or more comments from the companion pet’s perspective (e.g., “I love the rain because I get to play in the mud, even though Mom and Dad hate it. My favorite thing to do is...”), or both. Additionally, a colorful circle may be displayed around the image of the companion pet. The colorful circle (described in more detail below) may correspond to the breed makeup of the companion pet.

[0019] FIG. 1B further illustrates a breed field at the top of the "About" page in an exemplary environment of a platform for inferring companion pet kinship, according to one or more embodiments.

[0020] A breed section at the top of the Overview page can be accessed by scrolling down (e.g., swiping up) from the section shown in FIG. 1A. In the breed section at the top of the page, the platform can further display a breakdown of the breed composition of the companion pet. For example, the companion pet shown in FIG. 1B has the following breed composition: Great Pyrenees (26%), German Shepherd (15%), Beauceron (13%), White Swiss Shepherd (13%), and Chow Chow (13%). In addition, the platform can display only breeds with a percentage equal to or greater than a percentage threshold. The percentage threshold can be, for example, 6.25%. For example, the machine learning model may identify eight different breeds as breeds possessed by the companion pet. However, the machine learning model may also identify that three of these eight different breeds each account for less than 6.25% of the companion pet. In that case, the platform can display only the five breeds that are above the 6.25% threshold. Additionally, the platform may display a circle around the image of the companion pet that visually represents the breed makeup of the companion pet. For example, in FIG. 1B, the Great Pyrenees breed is assigned a specific color. Since 26% of companion pets are Great Pyrenees, the 26% portion of the circle has the specific color that represents Great Pyrenees. The circle may also visually represent a complete breakdown of all the breeds that make up the companion pet.

[0021] The platform may also display a total number of relatives of the companion pet. The total number of relatives may be determined using a machine learning model for inferring kinship relationships. Such machine learning models are described in more detail below. In addition to displaying the total number of relatives, the platform may also display a breakdown of the total number of relatives. More specifically, the platform may display the number of close relatives, the number of somewhat extended relatives, the number of distant relatives, or any combination thereof. For example, in FIG. 1B, the platform indicates that the companion pet has 8 close relatives, 12 somewhat extended relatives, and 43 distant relatives.

[0022] FIG. 1C further illustrates a close relatives section of the "Overview" page in an exemplary environment of a platform for inferring companion pet kinship, according to one or more embodiments.

[0023] The close relatives section of the Overview page can be accessed by scrolling further down (e.g., swiping up) from the section shown in FIG. 1B. In the close relatives section, the platform can display thumbnails of a portion of the profile thumbnails of companion pets that are matched as close relatives of the companion pet. For example, the platform can display only thumbnails of the profile thumbnails of companion pets that are matched as close relatives of the companion pet. Each profile thumbnail can include an image, a name, a circular ring around the image, a location, a kinship type, a commonality ratio, or any combination thereof. For example, the image can be a photo of the matched companion pet and a circular ring can be around the perimeter of the image. The circular ring can visually represent the breed makeup of the matched companion pet. The profile thumbnail can also include a name (e.g., Emma) and a location (e.g., Denver, Colorado). The platform may also display the type of kinship between the matched companion pet and the companion pet (e.g., close relative). For example, the platform may display whether the matched companion pet is a close relative, a more distant relative, or a distant relative of the companion pet. The platform may also display a commonality ratio indicating the percentage of DNA fragments shared between the companion pet and its matched companion pet. For example, the commonality ratio may be expressed as a percentage (e.g., 50%). The matched companion pets may be ordered in the close relatives section based on the commonality ratio. Additionally, the platform may present the user with an option to view all matched companion pets (e.g., "View all 63 relatives") rather than just a subset of the matched companion pets.

[0024] 1D illustrates a "Relatives" page in an exemplary environment of a platform for inferring kinship of companion pets, according to one or more embodiments. For example, when a user selects a "See all 63 relatives" button on the "Overview" page, the platform may display all profile thumbnails corresponding to matched companion pets. The platform may display relationship labels (e.g., mother) on some and / or all of the profile thumbnails.

[0025] FIG. 1E illustrates a detailed profile page of a matched companion pet in an exemplary environment of a platform for inferring companion pet kinship according to one or more embodiments.

[0026] For example, when a user selects a profile thumbnail, the platform may display a detailed profile page corresponding to the matched companion pet represented by the profile thumbnail. The detailed profile page may include the name of the matched companion pet (e.g., Emma), an image of the matched companion pet, the age of the matched companion pet (e.g., 3 years and 1 month old), the gender of the matched companion pet (e.g., female), one or more comments from the perspective of the matched companion pet (e.g., "I love the rain because I get to play in the mud, even though Mom and Dad hate it. My favorite thing to do is..."), or any combination thereof.

[0027] FIG. 1F further illustrates a “Relationship” section of a matched companion pet's detailed profile page in an exemplary environment of a platform for inferring companion pet kinship, according to one or more embodiments.

[0028] A "Relationship" section on the matched companion pet's detailed profile page can be accessed by scrolling down (e.g., swiping up) from the section shown in FIG. 1E. The platform can also further display kinship, infographics, commonality ratios, or any combination thereof. The platform can display the kinship between the companion pet and its corresponding matched companion pet. The infographic can provide a visual representation of the kinship. Additionally, the commonality ratio can indicate the percentage of DNA fragments shared between the companion pet and its matched companion pet. Additionally, the commonality ratio can be further illustrated using an infographic, such as a Venn diagram. Kinship can be inferred by machine learning models, a process that is described below.

[0029] 1G further illustrates a "Location" section of the matched companion pet's detailed profile page in an exemplary environment of a platform for inferring companion pet kinship, according to one or more embodiments. The "Location" section of the matched companion pet's detailed profile page can be accessed by scrolling down (e.g., swiping up) from the section shown in FIG. 1F. The platform can also display the matched companion pet's address (e.g., "Emma lives in Denver, Colorado") as well as a detailed map view displaying both the companion pet's address and the matched companion pet's address.

[0030] FIG. 1H illustrates a “Breed Comparison” section of a matched companion pet's detailed profile page in an exemplary environment of a platform for inferring companion pet kinship according to one or more embodiments.

[0031] A "Breed Comparison" section of the matched companion pet's detailed profile page can be accessed by scrolling down (e.g., swiping up) from the section shown in FIG. 1G. The platform can display a detailed comparison of the overlapping DNA fragment similarities between the companion pet and its matched companion pet. The detailed comparison can include a side-by-side comparison of the companion pet and its matched companion pet, a list of all breeds of at least one of the companion pet and its matched companion pet, and the like. The side-by-side comparison can also include side-by-side images of the companion pet and the matched companion pet, with circles surrounding each image. The side-by-side comparison can also include a ratio (e.g., percentage ratio) of how much of each of the listed breeds is present in the companion pet and its matched companion pet. Additionally, the platform may also display the number of breeds that the companion pet and its matched companion pet have in common (e.g., "Coco and Emma have 3 breeds in common.").

[0032] FIG. 1I illustrates a “Common Breeds” section of a detailed profile page of a matched companion pet in an exemplary environment of a platform for inferring companion pet kinship according to one or more embodiments.

[0033] A "Common Breeds" section of the matched companion pet's detail profile page may be accessed by scrolling down (e.g., swiping up) from the section shown in FIG. 1H. The platform may display a list of companion pet breeds that the companion pet and its matched companion pet have in common. This may be done by displaying a graphical image corresponding to each companion pet breed. Additionally, the platform may display a section or link where the user can obtain additional information about each common breed (e.g., "Great Pyrenees").

[0034] FIG. 1J illustrates a “Health” section of a matched companion pet's detailed profile page in an exemplary environment of a platform for inferring companion pet kinship, according to one or more embodiments.

[0035] A "Health" section of the matched companion pet's detailed profile page may be accessed by scrolling down (e.g., swiping up) from the section shown in FIG. 1I. The "Health" section may include information about health-related genes that the companion pet and its matched companion pet have in common. The platform may show the name of the health-related gene, the state of the health-related gene, the copy number of the health-related gene, an opportunity for the user to learn more about the health-related gene, or any combination thereof. For example, the name of the health-related gene may be "Degenerative Myelopathy (DM)" and the platform may display the number of copies of "Degenerative Myelopathy" that the companion pet and its matched companion pet have. Additionally, the platform may show the state of the health-related gene, indicating whether the companion pet and its matched companion pet are "At Risk" for the health-related gene or "Notable" for the health-related gene. The platform may also provide opportunities to learn more, such as a link to more information about "Degenerative Myelopathy" (e.g., "Learn about DM").

[0036] Training a machine learning model to infer kinship relationships FIG. 2 illustrates an exemplary process for training a machine learning model to infer kinship of companion pets, according to one or more embodiments. Initially, the method can include receiving DNA information of a companion pet (step 202). The DNA information of the companion pet can include at least one DNA sequence, at least one matched companion pet, and at least one relationship label corresponding to the kinship between the companion pet and the at least one matched companion pet. In some embodiments, the companion pet DNA information can further include at least one companion pet breed label (e.g., Schnauzer, Doberman, etc.). In addition, the DNA information can further include, for example, whether one or both of the two sets of chromosomes of the companion pet DNA information have a match with the at least one matched companion pet, and the number of DNA fragments that match the at least one matched companion pet. Additionally, at least one relationship label can include at least one of mother, father, sister, brother, uncle, aunt, niece, nephew, cousin, grandfather, grandmother, grandson, granddaughter, great-grandfather, great-grandmother, great-grandson, great-granddaughter, half-blood mother, half-blood father, half-blood sister, half-blood brother, half-blood uncle, half-blood aunt, half-blood niece, half-blood nephew, half-blood cousin, half-blood grandfather, half-blood grandmother, half-blood grandson, half-blood granddaughter, half-blood great-grandfather, half-blood great-grandmother, half-blood great-grandson, and half-blood great-granddaughter.

[0037] The method may include training a machine learning model (step 204) to infer at least one kinship relationship from at least one DNA sequence upon receiving the companion pet DNA information, as described in more detail below in the discussion of steps 206, 208, and 210. In one embodiment, the machine learning model may be a decision tree model for multi-label classification (e.g., gradient boosting of decision trees for multi-label classification). However, the machine learning model contemplated in this disclosure is not limited to a decision tree model, but may be any model configured to solve the classification problem presented in this disclosure.

[0038] The training step can include extracting at least one local DNA fragment pattern from a database for each of the at least one DNA sequence, corresponding to a matched companion pet (step 206). The DNA sequence can be input into a database of companion pet DNA sequences. The database can contain hundreds, thousands, or millions of companion pet DNA sequence entries. The database can retrieve and output at least one local DNA fragment pattern, which can include at least one locally matching chromosome segment. The locally matching chromosome segment can be a perfect match to one or more chromosome segments of the input DNA sequence. Upon receiving the DNA sequence, the database can provide a narrowing option, such as a narrowing option by the length of the locally matching portion. For example, it can be useful to narrow the locally matching portion to 18 centimorgans in length to eliminate short identical segments that may be matched in the majority of companion pet breeds as a result of inbreeding. To account for possible errors in DNA genotyping and phasing analysis, the resulting local chromosome segments can be stitched together, and these stitched local chromosome segments can be collected for all chromosomes.

[0039] The training step may further include identifying at least one inferred kinship relationship between the companion pet and its matched companion pet by analyzing at least one local DNA fragment pattern (step 208). For example, the at least one inferred kinship relationship may include at least one of: mother, father, sister, brother, uncle, aunt, niece, nephew, cousin, grandfather, grandmother, grandson, granddaughter, great-grandfather, great-grandmother, great-grandson, great-granddaughter, half-mother, half-father, half-sister, half-brother, half-uncle, half-aunt, half-niece, half-nephew, half-cousin, half-grandfather, half-grandmother, half-grandson, half-granddaughter, half-great-grandfather, half-great-grandmother, half-great-grandson, and half-great-granddaughter. This analyzing step is described in more detail in FIG. 3 (below).

[0040] The training step may also include identifying at least one inferred kinship relationship and at least one corresponding confidence level based on the analyzing step (step 210). Additionally, the method may include determining a length of the match, where the length of the match corresponds to a number of fragments in the at least one local DNA fragment pattern. The method may further include storing the at least one inferred kinship relationship, the at least one corresponding confidence level, or both in a result database. The method may also include displaying the at least one inferred kinship relationship, the at least one corresponding confidence level, the length of the match, or any combination thereof, to a user via a user interface. An example of a user interface displaying two inferred kinship relationships (e.g., "mother" and "father") is shown in FIG. 1C and FIG. 1D.

[0041] It should be noted that while Figure 2 illustrates example blocks of the example method 200, in some implementations example method 200 may include additional blocks, fewer blocks, different blocks, or a different order of blocks than those illustrated in Figure 2. Additionally or alternatively, two or more blocks of example method 200 may be performed simultaneously.

[0042] 3 may include an exemplary method 300 for training a machine learning model to identify at least one inferred kinship relationship between a companion pet and at least one matched companion pet by analyzing at least one local DNA fragment pattern, the method 300 shown in FIG. 3 corresponding to step 208 of FIG. 2.

[0043] As discussed above, the method can include identifying at least one inferred kinship relationship between the companion pet and at least one matched companion pet by analyzing at least one local DNA fragment pattern (step 302). Details of this analyzing step or substeps are described in more detail below with reference to steps 304-314. Note that one or more of the indicators identified in steps 304-314 can be used to train a machine learning model.

[0044] At least one local DNA fragment pattern can include multiple homoeologous (IBD) fragments. A homoeologous fragment can be a long stretch of DNA fragment that is identical in sequence between a companion pet and its matched companion pet. However, not all DNA fragments that are identical between two companion pets are considered homoeologous fragments. For example, purebred dogs are inbred, so there may be long stretches of identical DNA between dogs of that lineage.

[0045] The analyzing step may further include analyzing a plurality of homoeologous fragments to identify a total number of homoeologous fragments contained in at least one local DNA fragment pattern (step 304).

[0046] The analyzing step may also include analyzing the plurality of homoeologous fragments to determine a combined length of the plurality of homoeologous fragments in at least one local DNA fragment pattern (step 306), where the combined length of the plurality of homoeologous fragments may be measured in centimorgans.

[0047] Additionally, the analyzing step may further include a step (step 308) of analyzing at least one local DNA fragment pattern to determine whether the at least one local DNA fragment pattern occurs in one or both of the two sets of chromosomes possessed by the companion pet.

[0048] The analyzing step may further include a step of determining a proportion of the at least one DNA sequence that does not contain any of the multiple homoeologous fragments (step 310). The analyzing step may further include a step of determining a proportion of the at least one DNA sequence that contains multiple homoeologous fragments only in one of the two chromosomes (step 312). The analyzing step may further include a step of determining a proportion of the at least one DNA sequence that contains multiple homoeologous fragments in both of the two chromosomes (step 314).

[0049] The above analysis (e.g., various indicators identified during the analysis) can be used to assess the kinship between a companion pet and its matched companion pet (by providing these indicators as features to a machine learning model). For example, identifying the proportion of at least one DNA sequence that does not contain any of the multiple homoeologous fragments (called IBD0) can be useful in determining whether or not there is a possible kinship. If the proportion of IBD0 is very high, it is considered that there is a low possibility of kinship. Furthermore, taking the example of two half-brothers (e.g., with one parent), such two animals inherit a pair of chromosomes from the same parent. Thus, the two half-brothers and sisters will only be homoeologous at the pair of chromosomes inherited from the same parent. Such a situation (called IBD1) can be demonstrated by identifying the proportion of DNA sequences in both the companion pet and its matched companion pet that have homoeologous fragments in only one of the two pairs of chromosomes. As another example, take the example of full-blood (e.g., same-parent) siblings, in which case both of the two pairs of chromosomes are homoeologous between the full-blood siblings because both pairs of chromosomes share the same ancestry. This situation (called IBD2) can be demonstrated by determining the proportion of DNA sequences in both companion pets and their matched companion pets that are homoeologous in both pairs of chromosomes. Examples of such situations are further illustrated in Table 1 (below).

[0050] [Table 1]

[0051] The method may also include a step of identifying a level of homozygosity-by-descent (HBD) (homozygosity) for the companion pet and at least one matched companion pet (step 316). The homozygosity provides information about the degree of inbreeding in the companion pet's lineage. Purebred animals are expected to have a higher degree of homozygosity than outbred animals. For example, for each pair of companion pets and their matched companion pets, the two indices of the homozygosity of the companion pet and the homozygosity of the matched companion pet are used as additional indices in the overall analysis performed by the machine learning model. In other words, in addition to the indices identified in steps 304-314, the homozygosity of the companion pet and the homozygosity of the matched companion pet can also be used to train the machine learning model.

[0052] In one embodiment, cross-validation can be used to determine the accuracy of the model, for example by masking a portion of the ground truth relationship labels in the training dataset (i.e., the training panel) and estimating the masked labels based on a machine learning model built from the remainder of the training dataset.

[0053] It should be noted that while Figure 3 illustrates example blocks of the example method 300, in some implementations the example method 300 may include additional blocks, fewer blocks, different blocks, or a different order of blocks than those illustrated in Figure 3. Additionally or alternatively, two or more blocks of the example method 300 may be performed simultaneously.

[0054] FIG. 4 further illustrates a method 400 for training a machine learning model to infer at least one shared trait between a companion pet and a matched companion pet, according to one or more embodiments.

[0055] The method may include identifying at least one overlapping trait in at least one local DNA fragment pattern (step 402). For example, the at least one trait may be a breed specific trait. Examples of traits may include eye color, coat color, and health-related genes. Upon identifying at least one overlapping trait, the method may further include outputting a notification of the at least one shared trait with at least one matched companion pet (step 404).

[0056] It should be noted that while Figure 4 illustrates example blocks of the example method 400, in some implementations the example method 400 may include additional blocks, fewer blocks, different blocks, or a different order of blocks than those illustrated in Figure 4. Additionally or alternatively, two or more blocks of the example method 400 may be performed simultaneously.

[0057] 5 further illustrates a method 500 for training a machine learning model to output a list of DNA-matched companion pet samples, according to one or more embodiments, where method 500 may be complementary to method 200 shown in FIG. 2, in which step 502 corresponds to step 210 of FIG.

[0058] As described above in the description of FIG. 2, the method may include identifying at least one inferred kinship relationship and at least one corresponding confidence level based on the analyzing step (step 502). The method may further include outputting a list of DNA-matched companion pet samples based on the identifying step performed by the machine learning model (step 504). Here, the DNA-matched companion pet samples are companion pet samples that meet or exceed a match threshold. For example, when two or more local DNA fragment patterns are returned from the database, all of the returned local DNA fragment patterns may be displayed in a list. Also, only the local DNA fragment patterns that meet or exceed a threshold may be displayed in a list. The threshold may include a maximum number of items to display (e.g., 50 items) and may include a cutoff DNA match percentage value. The cutoff DNA match percentage value may be calculated by determining a percentage of local DNA fragments in the database that match all of the DNA fragments of the companion pet. For example, the thresholds may include a cutoff DNA percentage value of 6.25%.

[0059] It should be noted that while Figure 5 illustrates example blocks of the example method 500, in some implementations the example method 500 may include additional blocks, fewer blocks, different blocks, or a different order of blocks than those illustrated in Figure 5. Additionally or alternatively, two or more blocks of the example method 500 may be performed simultaneously.

[0060] Inferring kinship using trained machine learning models Described below is an exemplary process for inferring kinship among companion pets (e.g., by utilizing a trained machine learning model, such as a machine learning model trained according to one or more embodiments described above), in accordance with one or more embodiments.

[0061] The method can include receiving DNA information of the companion pet. The DNA information of the companion pet can include at least one DNA sequence. The receiving of the DNA information of the companion pet can be performed by a machine learning model. The machine learning model can be pre-trained.

[0062] Upon receiving the companion pet DNA information, the method can include using a trained machine learning model to infer at least one kinship relationship from the at least one DNA sequence, as described in more detail below.

[0063] The method may include extracting at least one local DNA fragment pattern from a database for each of the at least one DNA sequence, the local DNA fragment pattern corresponding to at least one matched companion pet. The DNA sequence may be input into a companion pet DNA sequence database. The DNA sequence database may contain hundreds, thousands, or millions of companion pet DNA sequence entries. Upon receiving the DNA sequence, the DNA sequence database may search for and output at least one local DNA fragment pattern, the at least one local DNA fragment pattern may include at least one locally matching chromosome segment. The local chromosome segment may be an exact match to one or more chromosome segments of the input DNA sequence. Upon receiving the DNA sequence, the database may provide a refinement option, such as a refinement option by the length of the locally matching portion. For example, it may be useful to refine the locally matching portion to 18 centimorgans in length to eliminate short identical segments that may be matched as a result of inbreeding. To account for possible errors in DNA genotyping and phasing analysis, the resulting local chromosome segments can be stitched together, and these stitched local chromosome segments can be collected for all chromosomes.

[0064] The method may include identifying at least one inferred kinship relationship between the companion pet and at least one matched companion pet by analyzing at least one local DNA fragment pattern. For example, the at least one inferred kinship relationship may include at least one of mother, father, sister, brother, uncle, aunt, niece, nephew, cousin, grandfather, grandmother, grandson, granddaughter, great-grandfather, great-grandmother, great-grandson, great-granddaughter, half-mother, half-father, half-sister, half-brother, half-uncle, half-aunt, half-niece, half-nephew, half-cousin, half-grandfather, half-grandmother, half-grandson, half-granddaughter, half-great-grandfather, half-great-grandmother, half-great-grandson, and half-great-granddaughter. The analyzing step is described in more detail below.

[0065] At least one local DNA fragment pattern can contain multiple homoeologous fragments, but not all identical DNA fragments between two companion pets are considered homoeologous fragments (for example, purebred pets are inbred, which may result in long stretches of identical DNA between pets of the same lineage).

[0066] The analyzing step may further include analyzing a plurality of homoeologous fragments to identify a total number of homoeologous fragments included in at least one local DNA fragment pattern.

[0067] The analyzing step can also include analyzing a plurality of homoeologous fragments to determine a combined length of the plurality of homoeologous fragments in at least one local DNA fragment pattern, the combined length of the plurality of homoeologous fragments being measured in centimorgans.

[0068] Additionally, the analyzing step may further include a step of analyzing at least one local DNA fragment pattern to determine whether the at least one local DNA fragment pattern occurs in one or both of the two sets of chromosomes possessed by the companion pet.

[0069] The analyzing step may further include a step of determining the proportion of DNA sequences that do not contain any of the multiple homoeologous fragments among the at least one DNA sequence. The analyzing step may further include a step of determining the proportion of DNA sequences that contain multiple homoeologous fragments only in one of the two pairs of chromosomes among the at least one DNA sequence. The analyzing step may further include a step of determining the proportion of DNA sequences that contain multiple homoeologous fragments in both of the two pairs of chromosomes among the at least one DNA sequence.

[0070] The above analysis (e.g., various indices identified during the analysis) can be used to assess the kinship between a companion pet and its matched companion pet (by providing the identified indices as features to a trained machine learning model). For example, identifying the proportion of at least one DNA sequence that does not contain any of the multiple homoeologous fragments (called IBD0) can be useful in determining whether or not there is a possible kinship. If the proportion of IBD0 is very high, it is considered that there is a low possibility of kinship. Furthermore, taking the example of two half-blood (e.g., one parent) siblings, such two animals inherit a pair of chromosomes from the same parent. Thus, the two half-blood siblings will only be homoeologous at the pair of chromosomes inherited from the same parent. Such a situation (called IBD1) can be demonstrated by identifying the proportion of DNA sequences in both the companion pet and its matched companion pet that are homoeologous at only one pair of chromosomes out of the two pairs. As another example, in the case of full-blood siblings (e.g., with the same parents), both pairs of chromosomes are homoeologous between full-blood siblings because both pairs of chromosomes share the same ancestry. This situation (called IBD2) can be demonstrated by determining the proportion of DNA sequences in both companion pets and their matched companion pets that are homoeologous between both pairs of chromosomes.

[0071] The method may also include a step of determining a homozygosity for the companion pet and at least one of the matched companion pets, which indicates the degree of inbreeding. The homozygosity provides information about the degree of inbreeding in the lineage of the companion pet. Purebred animals are expected to have a higher degree of homozygosity than outbred animals. For example, for each pair of companion pets and matched companion pets, the two indices of the homozygosity of the companion pet and the homozygosity of the matched companion pet are used as additional indices in the global analysis performed by the machine learning model. In other words, in addition to the indices previously determined (e.g., by analyzing at least one local DNA fragment pattern), the homozygosity of the companion pet and the homozygosity of the matched companion pet can also be provided to the trained machine learning model for inferring kinship.

[0072] The method can also include identifying at least one overlapping trait in the at least one local DNA fragment pattern. For example, the at least one trait can be a breed specific trait. Examples of traits can include eye color, coat color, and health-related genes. Upon identifying the at least one overlapping trait, the method can also include outputting a notification indicating the at least one trait shared with the at least one matched companion pet.

[0073] Based on the analysis performed by the machine learning model, at least one inferred kinship relationship between the companion pet and the matched companion pet and at least one corresponding confidence level can be identified. Additionally, the method can further include determining a length of the match, where the length of the match corresponds to a number of fragments in the at least one local DNA fragment pattern. The method can further include storing the at least one inferred kinship relationship, the at least one corresponding confidence level, or both in a results database. The method can also include displaying the at least one inferred kinship relationship, the at least one corresponding confidence level, the length of the match, or any combination thereof, to a user via a user interface. An example user interface displaying two inferred kinship relationships (e.g., "mother" and "father") is shown in FIG. 1C and FIG. 1D.

[0074] The method may further include outputting a list of DNA-matched companion pet samples based on the identifying step performed by the machine learning model. Here, the DNA-matched companion pet samples are companion pet samples that meet or exceed a match threshold. For example, if two or more local DNA fragment patterns are returned from the database, all of the returned local DNA fragment patterns may be displayed in a list. Also, only the local DNA fragment patterns that meet or exceed a threshold may be displayed in a list. The threshold may include a maximum number of items to display (e.g., 50 items) and may include a cutoff DNA match percentage value. The percentage value may be calculated by identifying a percentage of the local DNA fragments of the companion pets in the database that match all of the DNA fragments of the companion pets. For example, the threshold may include a cutoff DNA percentage value of 6.25%.

[0075] An exemplary process for inferring kinship among companion pets FIG. 6 illustrates an exemplary process for inferring companion pet kinship using companion pet DNA information according to one or more embodiments.

[0076] The method can include receiving DNA information of a companion pet, step 602. The DNA information of the companion pet can include at least one DNA sequence.

[0077] The method may include transmitting at least one DNA sequence to a database (step 604). The database may receive the at least one DNA sequence and search for and output at least one local DNA fragment pattern, the at least one local DNA fragment pattern including at least one matching chromosome segment corresponding to the at least one matched companion pet. The DNA sequence may be input to a database of companion pet DNA sequences. The database may contain hundreds, thousands, or millions of companion pet DNA sequence entries. The database may receive the DNA sequence and search for and output at least one local DNA fragment pattern, the at least one local DNA fragment pattern including at least one locally matching chromosome segment, the locally matching chromosome segment being an exact match to one or more chromosome segments of the input DNA sequence. Upon receiving the DNA sequence, the database may provide refinement options, such as by length of the locally matching portion. For example, it may be useful to narrow the length of local matches to 18 centimorgans to eliminate short identical segments that may be the result of inbreeding across a majority of companion pet breeds. The resulting local chromosome segments can also be stitched together to account for possible errors in DNA genotyping and phasing analysis. These stitched local chromosome segments can then be collected for all chromosomes.

[0078] The method can include identifying at least one inferred kinship relationship between the companion pet and at least one matched companion pet by analyzing at least one local DNA fragment pattern (step 606). The at least one local DNA fragment pattern can include multiple identical by descent (IBD) fragments.

[0079] The method may also include outputting at least one inferred kinship relationship based on the analyzing step (step 608). The at least one inferred kinship relationship may include at least one of mother, father, sister, brother, uncle, aunt, niece, nephew, cousin, grandfather, grandmother, grandson, granddaughter, great-grandfather, great-grandmother, great-grandson, great-granddaughter, half-mother, half-father, half-sister, half-brother, half-uncle, half-aunt, half-niece, half-nephew, half-cousin, half-grandfather, half-grandmother, half-grandson, half-granddaughter, half-great-grandfather, half-great-grandmother, half-great-grandson, and half-great-granddaughter. The method may also include identifying at least one corresponding confidence level. Additionally, the method may include determining a length of the match, where the length of the match corresponds to a number of fragments in the at least one local DNA fragment pattern. The method may further include storing the at least one inferred kinship relationship, the corresponding at least one confidence level, or both in a results database. The method may also include displaying the at least one inferred kinship relationship, the corresponding at least one confidence level, the length of the match, or any combination thereof, to a user via a user interface. An example user interface displaying two inferred kinship relationships (e.g., "mother" and "father") is shown in Figures 1C and 1D.

[0080] In some embodiments, the method can include outputting a list of DNA-matched companion pet samples, where the DNA-matched companion pet samples meet or exceed a match threshold. For example, if more than one local DNA fragment pattern is returned from the database, all of the returned local DNA fragment patterns can be displayed in the list. Alternatively, only the returned local DNA fragment patterns that meet or exceed a threshold can be displayed in the list. The threshold can include a maximum number of results to display (e.g., 50 results) and can include a cutoff DNA match percentage value. The cutoff DNA match percentage value can be calculated by determining the percentage of local DNA fragments in the database that match all of the DNA fragments of the companion pet. For example, the threshold can include a cutoff DNA percentage value of 6.25%.

[0081] In some embodiments, the method can include identifying at least one overlapping trait in the at least one local DNA fragment pattern, such as eye color, coat color, or health-related genes. The method can also include outputting a notification indicating the at least one trait is shared with at least one matched companion pet, where the at least one trait can be a breed-specific trait.

[0082] In some embodiments, the method may also include determining a homozygosity for the companion pet and at least one matched companion pet, which indicates the degree of inbreeding for each companion pet. The homozygosity may provide information regarding the degree of inbreeding in the companion pet's lineage. Purebred animals may be expected to have a higher homozygosity than outbred animals.

[0083] In some embodiments, the method can include analyzing a plurality of homoeologous fragments to identify a total number of homoeologous fragments contained in at least one local DNA fragment pattern.

[0084] In some embodiments, the method can further include analyzing the plurality of homoeologous fragments to determine a combined length of the plurality of homoeologous fragments in at least one local DNA fragment pattern, where the combined length of the plurality of homoeologous fragments can be measured in centimorgans.

[0085] In some embodiments, the method may further include analyzing at least one local DNA fragment pattern to determine whether the at least one local DNA fragment pattern occurs in one of the two pairs of chromosomes possessed by the companion pet, or in both of the two pairs of chromosomes. In some embodiments, the method may further include determining the proportion of DNA sequences in the at least one DNA sequence that do not contain any of the multiple homoeologous fragments. The analyzing step may further include determining the proportion of DNA sequences in the at least one DNA sequence that contain multiple homoeologous fragments only in one of the two pairs of chromosomes. The analyzing step may further include determining the proportion of DNA sequences in the at least one DNA sequence that contain multiple homoeologous fragments in both of the two pairs of chromosomes.

[0086] It should be noted that while Figure 6 illustrates example blocks of the example method 600, in some implementations the example method 600 may include additional blocks, fewer blocks, different blocks, or a different order of blocks than those illustrated in Figure 6. Additionally or alternatively, two or more blocks of the example method 600 may be performed simultaneously.

[0087] Exemplary Environment and Exemplary Device 7 illustrates an example environment 700 in which the technology presented herein can be utilized. One or more user devices 705, one or more external systems 710, and one or more server systems 715 can communicate over a network 701. As described in more detail below, the one or more server systems 715 can communicate with one or more other components of the environment 700 over the network 701. The one or more user devices 705 can be associated with a user, e.g., a user associated with one or more of generating, training, tuning a machine learning model for inferring kinship of companion pets.

[0088] In some embodiments, the components of environment 700 are associated with a public entity, such as, for example, a veterinarian, a clinic, an animal specialist, or a research center. In some embodiments, one or more of the components of environment 700 are associated with a different entity than the other components. The systems and devices of environment 700 may communicate in any configuration, including, among other things, to generate, train, and / or use machine learning models for inferring kinship among companion pets, as described below.

[0089] The user device 705 can be configured to enable a user to access and interact with other systems in the environment 700 through the user device 705. For example, the user device 705 can be a computer system such as a desktop computer, a mobile device, a tablet, etc. In some embodiments, the user device 705 can include one or more electronic applications (e.g., programs, plug-ins, browser extensions, etc.) installed on a memory of the user device 705.

[0090] The user device 705 may include a display / user interface (UI) 705A, a processor 705B, a memory 705C, a network interface 705D, or any combination thereof. The user device 705 may execute an operating system (O / S) and at least one electronic application, all of which are stored in the memory 705C, via the processor 705B. The electronic application may be a desktop program, a browser program, a web client, or a mobile application program (which may be a browser program in a mobile O / S), an applicant-specific program, system control software, system monitoring software, software development tools, and the like. For example, the environment 700 may extend the information of a web client accessible through a web browser. In some embodiments, the electronic application may be associated with one or more of the other components in the environment 700. The application may manage the memory 705C, such as a database, and send streaming data to the network 701. The display / UI 705A may be a display with a touch panel or other input system (e.g., mouse, keyboard, etc.) that allows a user to interact with the application and the O / S. Network interface 705D may be a TCP / IP network interface (e.g., for Ethernet or wireless communication with network 701). Processor 705B may generate data, receive user input from display / UI 705A, send messages to or receive messages from server system 715, or any combination thereof, in parallel with executing applications, and may further perform one or more steps before providing output to network 701.

[0091] The external system 710 can be, for example, one or more third party and / or auxiliary systems that can be integrated into or communicate with the server system 715 to perform various document information extraction tasks. The external system 710 can communicate with other devices or systems in the environment 700 via one or more networks 701. For example, the external system 710 can communicate with the server system 715 on one or more networks 701 via API (Application Programming Interface) access and can communicate with the user device 705 on one or more networks 701 via web browser access.

[0092] In various embodiments, the network 701 can be a wide area network ("WAN"), a local area network ("LAN"), a personal area network ("PAN"), or the like. In some embodiments, the network 701 includes the Internet, and the provision of information and data between various systems is online. The term "online" can mean connecting to or accessing source data or information from a location that is separate from other devices or networks connected to the Internet. Alternatively, the term "online" can refer to connecting to or accessing a network (wired or wireless) through a mobile communication network or device. The Internet is a global system of computer networks that allows a party in front of a networked computer or other device to obtain information from other computers and communicate with parties with other computers or devices. The most widely used part of the Internet is the World Wide Web (often abbreviated as "WWW" or simply "Web"). The term "website page" is intended to broadly encompass any location or data store that is made available online, for example by being hosted or operated by a computer system, and that may contain data that is configured to cause a program, such as a web browser, to perform operations such as receiving or transmitting data, processing, displaying a visual display, or generating an interactive interface.

[0093] Server system 715 may comprise an electronic data system and may comprise computer-readable memory, such as, for example, a hard drive, a flash drive, a disk, etc. In some embodiments, server system 715 provides and / or interacts with application programming interfaces for data exchange with other systems (e.g., one or more of the other components in the environment).

[0094] The server system 715 may include a database 715A and at least one server 715B. The server system 715 may be a computer, a system of computers (e.g., a rack server), a cloud service computer system, or any combination thereof. The server system may store or access the database 715A (e.g., hosted on a third party server or memory 715E). The server may include a display / UI 715C, a processor 715D, a memory 715E, a network interface 715F, or any combination thereof. The display / UI 715C may be a display with a touch panel or other input system (e.g., mouse, keyboard, etc.) that allows a person operating the server 715B to control the functions of the server 715B. The server system 715 may run an operating system (O / S) and at least one instance of a servlet program (both of which are stored in the memory 715E) via the processor 715D.

[0095] The server system 715 may be responsible for generating, storing, training, or using a machine learning model configured to infer kinship of companion pets. The server system 715 may include a machine learning model, instructions associated with the machine learning model, or both, such as instructions for generating the machine learning model, instructions for training the machine learning model, instructions for using the machine learning model, etc. The server system 715 may also include instructions for obtaining DNA information data (e.g., based on the output of the machine learning model), instructions for operating the display 715C to output kinship information data (e.g., adjusted based on the machine learning model), or both. The server system 715 may also include training data, such as at least one DNA sequence, at least one matched companion pet, and at least one relationship label corresponding to a kinship relationship between the companion pet and the at least one matched companion pet.

[0096] In some embodiments, a system or device other than the server system 715 is utilized to generate and / or train the machine learning model. For example, such a system may include instructions to generate the machine learning model, training data, and ground truth, instructions to train the machine learning model, or both. The resulting trained machine learning model may then be provided to the server system 715.

[0097] In general, a machine learning model includes a set of variables (e.g., nodes, neurons, filters, etc.) that are adjusted to different values ​​(e.g., by applying weights and biases to the variables) by applying training data. In supervised learning, for example, when the ground truth of the provided training data is known, training can begin by feeding the machine learning model samples of training data with the machine learning model's variables set to default values ​​(e.g., default values ​​randomly set based on Gaussian noise, pre-trained models, etc.). The output can then be compared to the ground truth to identify errors, which can then be back-propagated to the machine learning model to adjust the values ​​of the variables.

[0098] Training can be performed in any suitable manner (e.g., batch) and can include any suitable training method (e.g., stochastic gradient descent, non-stochastic gradient descent, gradient boosting, random forest, etc.). In some embodiments, a portion of the training data can be set aside during training, and the portion of the training data can be used to validate the trained machine learning model, e.g., to evaluate the accuracy of the trained model by comparing the output of the trained model for the portion of the training data to its ground truth. Training the machine learning model can be configured to cause the machine learning model to learn associations between DNA information data and kinship data. Thus, the trained machine learning model will be configured to respond to DNA information data by identifying kinship based on the associations learned by training.

[0099] In various embodiments, the variables of the machine learning model may be correlated with each other in any suitable relationship such that an output of the machine learning model is generated. For example, in some embodiments, the machine learning model may comprise a signal processing architecture configured to identify, isolate, or extract features, patterns, structures, or any combination thereof that the text has. For example, the machine learning model may comprise one or more convolutional neural networks ("CNN") configured to identify features in the document information data, and may further comprise additional architectures (e.g., binding layers, neural networks, etc.) configured to identify relationships between the identified features to identify locations in the document information data.

[0100] 7 illustrates the components of environment 700 as separate components, it should be understood that in some embodiments, the components of environment 700, or portions of the components, may be integrated with or incorporated into one or more other components. For example, a portion of display 715C may be integrated with user device 705, etc. In some embodiments, operations or aspects of one or more components described above may be distributed among one or more other components. Any suitable arrangement and integration of the various systems and devices of environment 700 may be used.

[0101] Further aspects of the machine learning model and the method of utilizing the machine learning model to infer kinship of companion pets are described in detail above in the method description. It should be noted that the method description above may describe components shown in FIG. 7 (e.g., server system 715, user device 705, or components therein) as performing various operations. However, it should be understood that in various embodiments, various components of the environment 700 described above may execute instructions or perform operations, including those described above. When a device performs an operation, a processor, actuator, etc. associated with the device may be considered to perform the operation. It should also be understood that various embodiments may add, omit, or rearrange steps in any suitable manner.

[0102] In general, any process or operation described in this disclosure that is understood to be computer-implementable, such as the processes illustrated in FIGS. 1A-6, may be performed by one or more processors of a computer system (e.g., any of the systems or devices of the environment 700 illustrated in FIG. 7), as described above. Operations (processes) or processing steps performed by one or more processors may also be referred to as operations. One or more processors may be configured to perform an operation by accessing instructions (e.g., software or computer readable code) that, when executed by one or more processors, cause the one or more processors to perform such an operation. These instructions may be stored in a memory of the computer system. The processor may be a central processing unit (CPU), a graphics processing unit (GPU), or any suitable type of processing device.

[0103] A computer system (e.g., a system or device that performs the processes or steps illustrated in the examples above) may include one or more computing devices (e.g., one or more of the systems or devices illustrated in FIG. 7). One or more processors of a computer system may be included in a single computing device or may be distributed across multiple computing devices. Additionally, the memory of the computer system may include memory of each of the multiple computing devices.

[0104] FIG. 8 is a simplified functional block diagram of a computer 800 that can be configured as a device for performing the methods shown in FIGS. 1A-6 according to an exemplary embodiment of the present disclosure. For example, the device 800 can include a central processing unit (CPU) 820. The CPU 820 can be any type of processor device, for example, any type of dedicated or general-purpose microprocessor device. As will be appreciated by those skilled in the art, the CPU 820 can also be a processor in a multi-core / multi-processor system operating alone or as a cluster of multiple computing devices operating as a cluster or server farm. The CPU 820 can also be connected to a data communication infrastructure 810 (e.g., a bus (BUS), a message queue, a network, or a multi-core message passing scheme).

[0105] The device 800 may also include a main memory 840 (e.g., random access memory (RAM)) and may further include a secondary memory 830. The secondary memory 830 (e.g., read only memory (ROM)) may be, for example, a hard disk drive or a removable storage drive. Such a removable storage drive may be, for example, a floppy disk drive, a magnetic tape drive, an optical disk drive, or a flash memory. The removable storage drive in this example reads and writes to a removable storage unit in a well-known manner. The removable storage unit may be, for example, a floppy disk, a magnetic tape, an optical disk, or the like, which may be read and written to by the removable storage drive. As will be appreciated by those skilled in the art, such a removable storage unit typically includes a computer usable storage medium having computer software, data, or both stored thereon.

[0106] In alternative embodiments, secondary memory 830 may comprise other similar means by which computer programs or other instructions can be loaded into device 800. Examples of such means include a program cartridge and cartridge interface (e.g., as found in video game devices) or a combination of a removable storage unit and interface that allows transfer of software and data from a removable storage unit to device 800, such as a removable memory chip and corresponding socket (e.g., EPROM or PROM).

[0107] Device 800 may also include a communications interface ("COM") 860. Communications interface 860 allows for the transfer of software and data between device 800 and external devices. Communications interface 860 may include a modem, a network interface (e.g., an Ethernet card), a communications port, a PCMCIA slot and card, and the like. The software and data transferred through communications interface 860 may be in the form of signals receivable by communications interface 860, such as electronic, electromagnetic, or optical signals. These signals may be provided to communications interface 860 via a communications path in device 800. Such a communications path may be implemented using, for example, a communications channel such as wire or cable, optical fiber, a telephone line, a cellular phone link, an RF link, and the like.

[0108] The hardware elements, operating systems, and programming languages ​​of such equipment are conventional in nature and are presumed to be fully familiar to those skilled in the art. The device 800 may also include input / output ports 850 for connecting input / output devices such as a keyboard, mouse, touch panel, monitor, display, etc. Of course, the various server functions may be distributed across multiple similar platforms to distribute the processing load. Alternatively, the server may be implemented by appropriate programming of a single computer hardware platform.

[0109] The programmatic aspects of the technology may be considered "products" or "articles of manufacture" representative of executable code and / or associated data carried or embodied by any type of machine-readable medium. The "storage" medium type of medium includes any or all of the tangible memory of a computer, processor, or the like, or its associated modules (e.g., various semiconductor memories, tape drives, disk drives, etc.), capable of providing non-transient storage locations for software programming at any time. All or part of the software may also be transmitted over various communications networks, such as the Internet. Such communications may, for example, allow the software to be loaded from one computer or processor to another, such as from a management server or host computer of a mobile communications network to load the software into a server's computer platform, or from a server to load the software into a mobile device. Other types of media on which software elements may be carried include light waves, radio waves, and electromagnetic waves (e.g., the light waves, radio waves, and electromagnetic waves used in physical interfaces between local devices, wired and optical fixed line networks, and various air links). Also, the physical elements that carry such waves (e.g., wired or wireless links, optical links, etc.) may be considered media carrying the software. As used herein, except when limited to non-transitory, tangible "storage" media, terms such as computer "readable medium" and machine "readable medium" refer to any medium that participates in providing instructions to a processor for execution.

[0110] Where specific activities are referred to in this disclosure, such description is for convenience only and is not intended to limit the disclosure. Those skilled in the art will recognize that the concepts underlying the disclosed devices and methods may be utilized in any activity suitable therefor. The present disclosure may be understood by reference to the above description and the accompanying drawings, in which like elements are designated by the same reference numerals.

[0111] The terms used in the above description may be interpreted in the broadest possible sense, even if the terms are used in describing a particular embodiment of the present disclosure in detail. In fact, some terms may be emphasized in the above description, but if a term is intended to be interpreted in a limited manner, the term is clearly and specifically defined in the above "Detailed Description of the Invention" section. Both the general description and the detailed description are merely illustrative of the present disclosure and do not limit the features described in the claims.

[0112] In this disclosure, the phrase "based on" means "based at least in part on." The singular forms "a," "an," and "the" are intended to include reference to the corresponding plurals unless the context clearly indicates otherwise. Additionally, the word "exemplary" is used in its "example" sense, not its "ideal" sense. The terms "comprise," "comprising," "include," "including," and other variations thereof are intended to have a non-exclusive inclusive meaning. Thus, when a process, method, or product "comprises" ("includes" or "has") recited elements, it does not necessarily include only the recited elements, but may include other elements not expressly recited or inherent to such process, method, article, or apparatus. Additionally, the term "or" is used disjunctively, and the phrase "at least one of A or B" includes (A), (B), (A and A), (A and B), etc. Relative terms such as "substantially" and "generally" are used to indicate that there may be a variation of ±10% from the stated or implied value.

[0113] As used herein, terms such as "user" are intended to broadly encompass one or more pet parents. Terms such as "pet" are intended to broadly encompass a user's pet, and the term "pet" may encompass multiple pets. Terms such as "provider" are intended to broadly encompass a pet care business.

[0114] As used herein, the term "machine-learning model" is intended to broadly encompass instructions, data, models, or any combination thereof configured to weight, bias, classify, analyze, or generate output from received inputs. Such output may include any suitable type of output, such as a classification of the inputs, an analysis based on the inputs, or a design, process, inference, or advice related to the inputs. Typically, machine-learning models / systems are trained using training data (e.g., empirical data, samples of input data, or both) that is input to the model to set, adjust, or modify one or more aspects of the model (e.g., weights, biases, classification or clustering criteria, etc.). How aspects of the machine-learning model operate on the inputs may be linear, parallel, through a network (e.g., a neural network), or through any suitable configuration.

[0115] Execution of the machine learning model may include deployment of one or more machine learning techniques, such as linear regression, logistic regression, random forest, gradient boosted machine (GBM), decision tree, gradient boosting of decision tree, deep learning, deep neural network, or any combination thereof. Supervised training, unsupervised training, or both may be performed. For example, supervised learning may include providing training data and a number of labels corresponding to the training data (e.g., as ground truth). Meanwhile, unsupervised approaches may include clustering, classification, etc. Also, K-means clustering (unsupervised) or K-nearest neighbors (supervised) may be used. K-nearest neighbors and unsupervised cluster techniques may also be used in combination. Furthermore, any suitable type of training may be employed, such as, for example, stochastic methods, gradient boosting, random seed, recursive methods, epoch-based or batch-based methods.

[0116] In the above description of the exemplary embodiments of the present invention, various features of the present invention may be grouped together in one embodiment, one figure, or description thereof, for the purpose of streamlining the disclosure and facilitating understanding of one or more of the various aspects of the present invention. However, such a method of disclosure should not be interpreted as reflecting an intention that the present invention requires more features than are expressly recited in each claim. Rather, as recited in the following claims, the number of features that provide an inventive step to the present invention is fewer than all of the features shown in the one embodiment disclosed above. Therefore, the claims following this Detailed Description of the Invention are expressly incorporated into the Detailed Description of the Invention, and each claim in the claims is intended to stand alone as an embodiment of the present invention.

[0117] Furthermore, while some of the embodiments described herein may include some features of other embodiments but not others, it is understood by those skilled in the art that combinations of features from different embodiments are within the scope of the present invention and are intended to form various different embodiments. For example, in the following claims, any of the embodiments defined by the claims may be used in any combination.

[0118] Thus, although specific embodiments have been described, those skilled in the art will recognize that other and further modifications may be made without departing from the spirit of the claims, and that all such modifications and variations are intended to be included within the scope of the present invention. For example, functions shown in the block diagrams may be added or removed, steps may be interchanged between functional blocks, etc. Also, steps may be added or removed from methods described within the scope of the present invention.

[0119] The subject matter of the present disclosure has been described above, but it should be considered as illustrative and not limiting, and the appended claims are intended to include all modifications, extensions, and other embodiments that fall within the true spirit and scope of the present disclosure. Thus, the scope of the present disclosure should be determined by interpreting the following claims and their equivalents in the broadest possible sense to the maximum extent permitted by law, and should not be limited or restricted by the detailed description of the invention set forth above. Although various embodiments of the present disclosure have been described above, it will be apparent to one skilled in the art that many more embodiments are possible within the scope of the present disclosure. Thus, the present disclosure is not limited except as limited by the claims and their equivalents. [Explanation of symbols]

[0120] 700 Environment 701 Network 705 User Devices 705A User Device Display / User Interface 705B User device processor 705C User Device Memory 705D USER DEVICE NETWORK INTERFACE 710 External Systems 715 Server System 715A Database 715B Server 715C Server System Display / User Interface 715D Server System Processor 715E Server System Memory 715F Server System Network Interface 800 devices 810 Data Communication Infrastructure 820 Central Processing Unit 830 Auxiliary Memory 840 Main memory 850 I / O ports 860 Communication Interface

Claims

1. 1. A computer-implemented method for inferring companion pet kinship using companion pet DNA information, comprising: receiving DNA information for a companion pet, the DNA information including at least one DNA sequence; transmitting the at least one DNA sequence to a database, the database receiving the at least one DNA sequence retrieving and outputting at least one local DNA fragment pattern including at least one matching chromosome segment corresponding to the at least one matched companion pet; determining at least one inferred kinship relationship between said companion pet and at least one of said matched companion pets by analyzing said at least one local DNA fragment pattern; reducing the at least one inferred kinship relationship based on the analyzing step; The method includes:

2. 2. The computer-implemented method of claim 1, wherein the at least one local DNA fragment pattern comprises multiple identically ancestral (IBD) fragments.

3. 3. The computer-implemented method of claim 2, further comprising identifying a proportion of said at least one DNA sequence that contains said plurality of homoeologous fragments in only one of two chromosomes.

4. The computer-implemented method of claim 2 , further comprising identifying a proportion of the at least one DNA sequence that contains the plurality of homoeologous fragments in both of the two sets of chromosomes.

5. 3. The computer-implemented method of claim 2, further comprising identifying a proportion of said at least one DNA sequence that does not contain any of said plurality of homoeologous fragments.

6. 3. The computer-implemented method of claim 2, further comprising analyzing the plurality of homoeologous fragments to identify a total number of homoeologous fragments contained in the at least one local DNA fragment pattern.

7. 3. The computer-implemented method of claim 2, further comprising analyzing the plurality of homoeologous fragments to determine a total length of the plurality of homoeologous fragments contained in the at least one local DNA fragment pattern.

8. 8. The computer-implemented method of claim 7, wherein the combined length of the plurality of homoeologous fragments is measured in centimorgans.

9. identifying at least one overlapping trait in said at least one local DNA fragment pattern; outputting a notification indicating that the at least one trait is shared between the at least one matched companion pet; The computer-implemented method of claim 1 , further comprising:

10. 10. The computer-implemented method of claim 9, wherein the at least one trait is a breed-specific trait.

11. 2. The computer-implemented method of claim 1, wherein the at least one inferred kinship relationship includes at least one of: mother, father, sister, brother, uncle, aunt, niece, nephew, cousin, grandfather, grandmother, grandson, granddaughter, great-grandfather, great-grandmother, great-grandson, great-granddaughter, half-blood mother, half-blood father, half-blood sister, half-blood brother, half-blood uncle, half-blood aunt, half-blood niece, half-blood nephew, half-blood cousin, half-blood grandfather, half-blood grandmother, half-blood grandson, half-blood granddaughter, half-blood great-grandfather, half-blood great-grandmother, half-blood great-grandson, and half-blood great-granddaughter.

12. The method further comprises outputting a list of DNA-matched companion pet samples; 2. The computer-implemented method of claim 1, wherein the DNA-matched companion pet samples are companion pet samples that meet or exceed a match threshold.

13. The computer-implemented method of claim 1 , further comprising identifying a degree of homozygosity for the companion pet and the matched at least one companion pet, the degree indicating a respective degree of inbreeding.

14. 2. The computer-implemented method of claim 1, further comprising analyzing the at least one local DNA fragment pattern to determine whether the at least one local DNA fragment pattern occurs in one or both of two sets of chromosomes possessed by the companion pet.

15. at least one memory for storing instructions; at least one processor configured to perform operations by executing the instructions; 1. A computer system for inferring kinship of companion pets using DNA information of companion pets, comprising: The operation, receiving DNA information for a companion pet, the DNA information including at least one DNA sequence; transmitting the at least one DNA sequence to a database, the database receiving the at least one DNA sequence retrieving and outputting at least one local DNA fragment pattern including at least one matching chromosome segment corresponding to the at least one matched companion pet; determining at least one inferred kinship relationship between said companion pet and at least one of said matched companion pets by analyzing said at least one local DNA fragment pattern; reducing the at least one inferred kinship relationship based on the analyzing step; 2. A computer system comprising:

16. 16. The computer system of claim 15, wherein the operations further include analyzing the at least one local DNA fragment pattern to determine whether the at least one local DNA fragment pattern occurs in one or both of two sets of chromosomes possessed by the companion pet.

17. 16. The computer system of claim 15, wherein the at least one inferred kinship relationship includes at least one of: mother, father, sister, brother, uncle, aunt, niece, nephew, cousin, grandfather, grandmother, grandson, granddaughter, great-grandfather, great-grandmother, great-grandson, great-granddaughter, half-mother, half-father, half-sister, half-brother, half-uncle, half-aunt, half-niece, half-nephew, half-cousin, half-grandfather, half-grandmother, half-grandson, half-granddaughter, half-great-grandfather, half-great-grandmother, half-great-grandson, and half-great-granddaughter.

18. A non-transitory computer-readable medium containing instructions, comprising: the instructions, when executed by a processor, cause the processor to perform operations to infer kinship of companion pets using DNA information of companion pets; The operation is receiving DNA information for a companion pet, the DNA information including at least one DNA sequence; transmitting the at least one DNA sequence to a database, the database receiving the at least one DNA sequence retrieving and outputting at least one local DNA fragment pattern including at least one matching chromosome segment corresponding to the at least one matched companion pet; determining at least one inferred kinship relationship between said companion pet and at least one of said matched companion pets by analyzing said at least one local DNA fragment pattern; reducing the at least one inferred kinship relationship based on the analyzing step; A non-transitory computer readable medium comprising:

19. 20. The non-transitory computer readable medium of claim 18, wherein the at least one inferred kinship relationship comprises at least one of: mother, father, sister, brother, uncle, aunt, niece, nephew, cousin, grandfather, grandmother, grandson, granddaughter, great-grandfather, great-grandmother, great-grandson, great-granddaughter, half-mother, half-father, half-sister, half-brother, half-uncle, half-aunt, half-niece, half-nephew, half-cousin, half-grandfather, half-grandmother, half-grandson, half-granddaughter, half-great-grandfather, half-great-grandmother, half-great-grandson, and half-great-granddaughter.

20. 20. The non-transitory computer readable medium of claim 18, wherein the operations further include analyzing the at least one local DNA fragment pattern to determine whether the at least one local DNA fragment pattern occurs in one or both of two sets of chromosomes possessed by the companion pet.

Images