Downsampling genomic sequence data

Abstract

A method for automatically downsampling DNA sequence data using a variational autoencoder and maintaining the genomic integrity of the original file is disclosed. The method includes: performing bootstrapping on the genomic sequence data by an encoder (142) to generate resampling; evaluating the non-representativeness and self-inconsistency of the resampling by the encoder (142), and selecting a representative resampling based on the evaluation; constructing a vector representation from genotype likelihood based on the selected representative sampling by a modified encoder (142); integrating mapping position information and genotype likelihood by an analysis engine to identify the optimal vector representation of the resampling; and decoding the identified optimal vector representation of the resampling by a modified decoder (144) to obtain a downsampled read file that is similar to the original file and maintains the genomic integrity of the original file.

Description

Background Art

[0001] The present invention relates generally to the field of genomic sequence data, and more particularly to downsampling deoxyribonucleic acid (DNA) sequence data.

[0002] Genomes are sequenced every day, where a single human genome may occupy 100 gigabytes of storage space, so the storage requirements for genome sequence data will grow from gigabytes to petabytes to exabytes. In addition, for every three billion bases of the human genome sequence, 30 times more data (about 100 gigabases) must be collected due to errors in sequencing, base determination, and genome alignment. This means that up to 2-40 exabytes of storage capacity will be required for the human genome in 2025. In digital signal processing, downsampling, compression, and decimation are terms associated with the resampling process in multi-rate digital signal processing systems. Both downsampling and decimation can be synonymous with compression, or they can describe the entire process of bandwidth reduction (filtering) and sampling rate reduction. When the process is performed on a sampled sequence or other continuous function of a particular signal, downsampling produces an approximation of the sequence obtained by sampling the signal at a lower rate (or density, as in the case of a photo).

[0003] The emergence of a massively parallel high-throughput sequencing technology called Next Generation Sequencing (NGS) has revolutionized the field of biological sciences. The emergence of NGS has led to an increase in the amount and availability of DNA sequence data. Ultra-high-scale and effective NGS sequencing machines can produce billions of short DNA reads exceeding terabytes of data in a single run. Short DNA reads are stored in FASTQ files (gigabytes of plain text files), which are files in FASTQ format, which are used for a wide range of applications, such as gene annotation, expression research, personalized treatment and precision medicine. The FASTQ format is a text-based format for storing biological sequences (typically nucleotide sequences) and their corresponding quality scores. For simplicity, sequence letters and quality scores are each encoded with a single American Standard Code for Information Interchange (ASCII) character. Summary of the invention

[0004] Embodiments of the present invention disclose a method, a computer program product, and a system for automatically downsampling DNA sequence data using a variational autoencoder while maintaining the genomic integrity of the original file, the method comprising: an encoder performing bootstrapping on genomic sequence data to generate resampling; the encoder evaluating the unrepresentativeness and self-inconsistency of the resampling, and selecting a representative resampling based on the evaluation; a modified encoder constructing a vector representation from genotype likelihood based on the selected representative sampling; an analysis engine integrating mapping position information and genotype likelihood to identify the optimal vector representation of the resampling; and a modified decoder decoding the identified optimal vector representation of the resampling to obtain a downsampled read file that is similar to the genomic integrity of the original file and maintains the genomic integrity of the original file. BRIEF DESCRIPTION OF THE DRAWINGS

[0005] Figure 1 is a functional block diagram illustrating a distributed data processing environment according to an embodiment of the present invention;

[0006] Figure 2A A block diagram of a downsampling component is shown according to an embodiment of the present invention. Figure 1 The method is executed on a server computer in a distributed data processing environment of the invention for downsampling DNA sequence data using a variational autoencoder;

[0007] Figure 2B is an example of an overview of a bootstrapping process according to an embodiment of the present invention;

[0008] Figure 2C The depiction of an embodiment of the present invention is shown in Figure 1 A block diagram of an overview of bioinformatics generation using a variational autoencoder for downsampling DNA sequence data within a distributed data processing environment of ;

[0009] Figure 3 The embodiment of the present invention is shown in Figure 1 The steps of operating a downsampling component on a computing device within a distributed data processing environment of the present invention, the downsampling component is used to automatically downsample DNA sequence data using a variational autoencoder; and

[0010] Figure 4 Describes an embodiment of the present invention. Figure 1 A block diagram of components of a server computer executing a downsampling component within a distributed data processing environment. DETAILED DESCRIPTION

[0011] Embodiments of the present invention recognize that there are problems with current methods and systems for genomic data analysis and integration and archiving of DNA sequence data. For example, high-throughput genome sequencing inherently produces files containing millions of reads per sample, files that require a large amount of computing resources for storage and processing, and samples that need to be efficiently aggregated (e.g., downsampled) to optimize computational efficiency. Embodiments of the present invention recognize that current downsampling methods cannot maintain the genome integrity of the original sample.

[0012] As is known in the art, the surge in genomic data has presented tremendous challenges in analyzing, integrating, sharing, and archiving DNA sequence data. Embodiments of the present invention address the above-mentioned problems by proposing a system and method for downsampling billions of reads in FASTQ files to reduce the file size and number of DNA reads while maintaining the integrity of the original information and avoiding the loss of important genomic information, wherein the loss of information is mainly attributed to the loss of genomic variation, such as structural variation encoded in the excess reads, thereby enabling effective and efficient downstream processing and analysis of genomic sequence data.

[0013] In addition, embodiments of the present invention use composite representation to summarize genomic information. Embodiments of the present invention consider genomic sequence data as a whole, and convert genomic information (such as genomic sequence data as a whole) into a representation (such as a numerical vector) that keeps the original biological information complete, wherein input data comes from the DNA sequencing of multiple readings (such as analyzing, identifying and / or explaining the conversion of multiple readings). In various embodiments of the present invention, the biological information mainly captured and encoded in the reading is different types of variation known in the art, such as insertion-deletion mutations (gains and losses (indel)), single nucleotide polymorphisms (SNPs) and / or coding number variations (CNVs), which can be used to explain different phenotypes. In addition, embodiments of the present invention utilize shallow sampling using bootstrap methods to calculate the minimum value of representativeness and consistency. In some embodiments of the present invention, the present invention is not focused on single nucleotide polymorphism (SNP) interpretation, but the genotype likelihood assessment of multiple positions in the reading. Embodiments of the present invention can use graphic Poisson distribution to estimate genotype likelihood. In various embodiments of the present invention, the present invention is a downsampling method, and is alignment-free in some cases. Embodiments of the present invention recognize that alignment works when the user (i.e., researcher) has a reference genome and that in most cases the reference genome is incomplete, resulting in alignment dependence on the reference genome (i.e., alignment is only as good as your reference). In various embodiments of the present invention, the present invention focuses on genotype likelihood assessments at multiple positions within a read to be used as prior information without relying on variant calling techniques.

[0014] The implementation of the embodiments of the present invention may take various forms, and will be described below with reference to the accompanying drawings (i.e., Figure 1-Figure 4 ) discusses exemplary implementation details.

[0015] Figure 1 is a functional block diagram illustrating a distributed data processing environment according to one embodiment of the present invention, generally designated 100. The term "distributed" as used in this specification describes a computer system that includes multiple physically distinct devices that operate together as a single computer system. Figure 1 Only an illustration of one implementation is provided and no limitation on the environments in which different embodiments may be implemented is implied. Many modifications to the described environment may be made by those skilled in the art without departing from the scope of the invention as recited in the claims. Distributed data processing environment 100 includes computing devices 110 and server computers 120 interconnected by network 130.

[0016] The network 130 may be, for example, a storage area network (SAN), a telecommunications network, a local area network (LAN), a wide area network (WAN) such as the Internet, wireless technology for exchanging data over short distances (using short wavelength ultra high frequency (UHF) radio waves in the industrial, scientific, and medical (ISM) band from 2.4 to 2.485 GHz from fixed and mobile devices, and a building personal area network (PAN), or a combination of the three), and may include wired, wireless, or fiber optic connections. The network 130 may include one or more wired and / or wireless networks that may receive and transmit data, voice, and / or video signals, including multimedia signals that include voice, data, text, and / or video data. In general, the network 130 may be a network that supports the computing device 110 and the server computer 120, as well as any other computing devices and / or storage devices within the distributed data processing environment 100 ( Figure 1 Any combination of connections and protocols for communication between (not shown).

[0017] In some embodiments of the present invention, computing device 110 may be, but is not limited to, a standalone device, a client, a server, a laptop computer, a tablet computer, a netbook computer, a personal computer (PC), a smart phone, a desktop computer, a smart TV, a smart watch, a radio, a stereo system, a cloud-based service (e.g., a cognitive cloud-based service), AR glasses, a virtual reality headset, any HUD known in the art, and / or any programmable electronic computing device capable of communicating with various components and devices within distributed data processing environment 100 via network 130 or any combination thereof. In general, computing device 110 may represent any programmable computing device or combination of programmable computing devices capable of executing machine-readable program instructions and communicating with users of other computing devices via network 130 and / or capable of executing machine-readable program instructions and communicating with server computer 120. In some embodiments, computing device 110 may represent multiple computing devices.

[0018] In some embodiments of the present invention, computing device 110 may represent any programmable electronic computing device or combination of programmable electronic computing devices capable of executing machine-readable program instructions, manipulating executable machine-readable instructions, and communicating with server computer 120 and other computing devices (not shown) within distributed data processing environment 100 via a network such as network 130. Computing device 110 may include an instance of user interface (interface) 106 and local storage 104. Figure 1 In various embodiments not depicted in FIG. 1 , computing device 110 may have multiple user interfaces. Figure 1 In other embodiments not shown in FIG. 1 , the distributed data processing environment 100 may include multiple computing devices, multiple server computers, and / or one of multiple networks. The computing device 110 may include internal and external hardware components, such as those described in detail in FIG. Figure 4 further depicted and described in detail.

[0019] User interface (interface) 106 provides an interface to prediction component 140. Computing device 110 can enable a user and / or client to interact with component 140 and / or server computer 120 in various ways via user interface 106, such as sending program instructions, receiving program instructions, sending and / or receiving messages, updating data, sending data, inputting data, editing data, collecting data, and / or receiving data. In one embodiment, interface 106 can be a graphical user interface (GUI) or a web user interface (WUI), and can display at least text, documents, web browser windows, user options, application interfaces, and operating instructions. Interface 106 can include data (e.g., graphics, text, and sound) presented to the user and control sequences used by the user to control operations. In another embodiment, interface 106 can be a mobile application software that provides an interface between a user of computing device 110 and server computer 120. Mobile application software or "app" can be designed to run on smart phones, tablet computers, and other computing devices. In one embodiment, interface 106 can enable a user of computing device 110 to at least send data, input data, edit data (annotate), collect data, and / or receive data.

[0020] Server computer 120 may be a stand-alone computing device, a management server, a web server, a mobile computing device, one or more client servers, or any other electronic device or computing system capable of receiving, sending, and processing data. In other embodiments, server computer 120 may represent a server computing system utilizing multiple computers, such as, but not limited to, a server system such as in a cloud computing environment. In another embodiment, server computer 120 may represent a computing system utilizing clustered computers and components (e.g., database server computers, application server computers, etc.) that act as a single seamless resource pool when accessed within distributed data processing environment 100. Server computer 120 may include internal and external hardware components, such as those described with reference to FIG. Figure 4 In some embodiments, server computer 120 may represent a plurality of server computers.

[0021] Each of the shared storage 124 and the local storage 104 may be a data / knowledge repository and / or database that may be written to and / or read by one or a combination of the components 140, the server computer 120, and the computing device 110. In the described embodiment, the shared storage 124 resides on the server computer 120, while the local storage 104 resides on the computing device 110. In another embodiment, the shared storage 124 and / or the local storage 104 may reside elsewhere in the distributed data processing environment 100, so long as each is accessible to and can be accessed by the computing device 110 and the server computer 120. The shared storage 124 and / or the local storage 104 may each be implemented with any type of storage device capable of storing data and configuration files that may be accessed and utilized by the server computer 120, such as, but not limited to, a database server, a hard drive, or flash memory.

[0022] In some embodiments of the present invention, the shared storage device 124 and / or the local storage device 104 can each be a hard drive, a memory card, a computer output to a laser disc (cold storage), and / or any form of data storage device known in the art. In some embodiments, the shared storage device 124 and / or the local storage device 104 can each be one or more cloud storage systems and / or databases linked to a cloud network. In various embodiments, the shared storage device 124 and / or the local storage device 104 can access, store and / or accommodate user data, physical room data and conference data and / or data shared throughout the distributed data processing environment 100.

[0023] In the depicted embodiment, the downsampling component (component) 140 executes on the server computer 120. In other embodiments, the downsampling component 140 may execute on the computing device 110. Figure 1 In various embodiments of the invention not shown in the drawings, downsampling component 140 may be executed on multiple server computers 120 and / or multiple computing devices 110. In some embodiments, component 140 may be located and / or executed anywhere within distributed data processing environment 100, so long as downsampling component 140 is connected to and / or in communication with computing devices 110 and / or server computers 120 via network 130. In the depicted embodiment, encoder 142, decoder 144, and analysis engine 146 are each located and executed on downsampling component 140. Figure 1In various embodiments not depicted in the drawings, the encoder 142, decoder 144, and analysis engine 146 may be executed anywhere within the distributed data processing environment 100 as long as the encoder 142, decoder 144, and / or analysis engine 146 are each connected to and / or in communication with the computing device 110 and / or the server computer 120 via the network 130.

[0024] In various embodiments of the invention, encoder 142 may be one or more of a modified encoder, an audio encoder, a simple encoder, a priority encoder, a compressor, a rotary encoder, a linear encoder, an absolute encoder, an incremental encoder, and / or any type of encoder known in the art. In various embodiments of the invention, decoder 144 may be one or more of an audio decoder, a binary decoder, a decompression decoder, an instruction decoder, an orthogonal decoder, a video decoder, and / or any other decoder known in the art. Analysis engine 146 may be one or more analysis engines as understood in the art.

[0025] In various embodiments, the downsampling component 140 utilizes a probabilistic resampling framework via an encoder 142 to select a subset of read fragments from the entire file. The encoder (e.g., encoder 142) performs a bootstrapped optimization to generate a resampled set for a given N size (e.g., N=1 million reads, which may have a total of 100 resamples). In various embodiments of the invention, the downsampling component 140 selects a subset of encoded read fragments from one or more resamples via an encoder 142 after satisfying one or more predetermined criteria. In various embodiments of the invention, a decoder 144 may be used to generate a final set of read fragments representing the original FASTQ file. In various embodiments of the invention, the decoder 144 is used to generate a final read fragment that can be placed in a final resample. In various embodiments of the invention, the decoder 144 may be used to generate x (i.e., the final representative resample). The decoder 144 may reconstruct one or more vector representations that encode the reads as they move through the pipeline to generate actual reads / DNA fragments.

[0026] In various embodiments of the present invention, the downsampling component 140 includes two steps: first, an encoder (e.g., encoder 142) retrieves and / or receives genomic sequence data and performs bootstrapping to generate resampling, wherein the non-representativeness and self-inconsistency of the generated resampling are evaluated, wherein the variation is minimized, and wherein the most representative resampling is selected; second, a decoder (e.g., decoder 144) receives the most representative sampling (e.g., the most representative resampling selected) as input, and generates a reconstruction of the original input reading. In various embodiments of the present invention, the downsampling component 140 removes duplicate readings from the FASTQ file to reduce the redundancy of genomic information. In various embodiments of the present invention, the downsampling component 140 can process FASTQ files containing millions and / or billions of readings by marking one or more readings (e.g., all readings) with a unique identifier and then mapping them to a sampled reference sequence. In various embodiments of the present invention, the downsampling component 140 can encode the position information and mapping of each reading fragment and the quality score through the encoder 142. In various embodiments of the invention, a statistical model is trained to estimate the genotype likelihood at each position, and this information is encoded for each read fragment by encoder 142.

[0027] In various embodiments of the present invention, the downsampling component 140 can resample the DNA fragments generated from the massively parallel sequencing without compromising (i.e., while maintaining) the genomic integrity of the original file. For example, genomic integrity refers to genomic variations that result in different phenotypes and are encoded in different reads (i.e., multiple or excessive reads). In this particular example, in order to maintain genomic integrity, the downsampling component 140 performs downsampling on one or more of the multiple reads, wherein the downsampling component 140 retains a majority or a predetermined amount of the reads.

[0028] In various embodiments of the invention, the downsampling component 140 can retrieve and / or receive genomic sequence data (e.g., DNA sequence data) from one or more users, one or more databases, and / or from one or more search engines via the encoder 142, and perform bootstrapping on the retrieved and / or received genomic sequence data to generate resamples. In various embodiments of the invention, the downsampling component 140 can perform probabilistic resampling via the encoder 142 to obtain a resampled set comprising a plurality of reads. In various embodiments of the invention, the component 140 can evaluate the non-representativeness and self-inconsistency of the resamples and select the best / "most representative" resample.

[0029] In various embodiments of the invention, the downsampling component 140 can construct a representation from the genotype likelihood (e.g., converting a genomic read in FASTQ format into a digital vector representation) via an encoder 142. The downsampling component 140 can integrate the mapping position information and the genotype likelihood via an analysis engine 146 to identify the optimal vector representation for resampling. In various embodiments of the invention, the downsampling component 140 can retrieve and / or receive the most representative sampling as input via a decoder 144, and generate a reconstruction of the original input read. In various embodiments of the invention, the downsampling component 140 can decode the optimal resampled vector representation via a decoder 144 to obtain a downsampled read file that is similar to and maintains the genome integrity of the original file.

[0030] In various embodiments of the invention, given several reads overlapping a genomic location, the algorithm estimates the genotype likelihood probability for each read by maximizing the parameters of the joint likelihood function. These parameters are based on the Poisson graphical distribution (PGD). The parameters θ0, ..., θ D-1 The likelihood of is the probability of sampling a reading from a distribution with these specific parameters. This can be simplified in Equation 1 as follows: 1Equation 1:

[0031]

[0032] If the likelihood value is calculated by combining the genotype counts of multiple reads, the detection statistics can remain the same. For each read, the probability of each possible likelihood genotype is estimated.

[0033] In various embodiments of the present invention, a bootstrap method is used to generate resamples. For each bootstrap iteration, given FASTQX N , where X is a FASTQ file with a total of N reads, a fraction of the reads is randomly selected to generate x 1n , sampling with replacement ( Figure 2B ). This process is repeated I times. In the next iteration of the resampling process x n Select by increasing the number of readouts.

[0034] To find the optimal Z i (i.e., Z, where Z is the optimal set of resamples obtained from the bootstrap), applying self-inconsistency and non-representativeness, where,

[0035] Self-inconsistency is defined using Equation 2:

[0036] 2Equation 2:

[0037]

[0038] The self-inconsistencies are then averaged using Equation 3:

[0039] 3Equation 3:

[0040]

[0041] Non-representativeness is defined using Equation 4:

[0042] Equation 4:

[0043]

[0044] The self-inconsistencies are then averaged using Equation 5:

[0045] Equation 5:

[0046]

[0047] Note that in Equation 1, N represents the sampling size (the number of reads in the original file), n represents the resampling size (the number of reads in the resampling), and θ represents the poison rate. Note that in Equations 2-5, Nr represents the number of resamplings, N represents the number of sequences / reads in each, K represents the k-mer size, and M represents the total number of samples. indicates non-representativeness, and Indicates self-inconsistency. In various embodiments of the invention, a set of resamples is selected for further processing by the analysis engine 146. In various implementations of the invention, the input to the analysis engine 146 is a set of resamples (e.g., 100 resamples, each containing 1 million readings (i.e., a selected set of 100 resamples, and each resample containing 1 million readings)). The analysis engine 146 can then process the 100 resamples to obtain the best resample that truly represents the original file. This is done by minimizing the steps in the equation as described below.

[0048] To obtain the best resample or set of resamples (ie, optimal resamples), where the best resample or set of resamples is the resample with the lowest values ​​of non-representativeness and self-inconsistency, the position information and genotype likelihood are combined and analyzed as follows:

[0049] Given in is the best resampling set obtained from bootstrapping, z i ……z n is a single resample with m reads, X is the entire FASTQ file and ), where y is the constraint obtained from the alignment, which includes the positional information of the reads and their corresponding genotype information minimizing: This means that [z, y] ≈ X, so that, taking into account one or more constraints, The optimal [z, y] is given at its minimum. In various embodiments of the invention, the optimal [z, y] is passed through a decoder 144, which produces a downsampled FASTQ file with a reduced number of reads, but the retained reads maintain the genomic integrity of the input file.

[0050] In other embodiments of the invention, the downsampling component 140 may also use one or more constraints to identify the optimal set of resamples (i.e., optimal Z). The one or more constraints that may be used by the downsampling component 140 to identify the optimal set of resamples may include: (i) selecting a maximum interval based on position information encoded from a mapping; (ii) calculating the distribution of intervals in each resample so that position coverage is maximized; (iii) finding the resample with the minimum overlap that has the maximum coverage distribution; (iv) calculating an average read weight based on the likelihood of a genotype occurring at each read; and / or calculating a total weight of the resample by taking into account the weights of the reads.

[0051] Figure 2A The diagram depicts Figure 1 A block diagram of a downsampling component 140 in communication with a computing device 110 and / or a server computer 120 within a distributed data processing environment 100 for downsampling DNA sequence data using a variational autoencoder. Figure 2A A specific example of an overview of the steps followed to downsample a FASTQ file using a modified variational autoencoder is provided. Figure 2A An illustration of an implementation is provided and does not imply any limitation on the environments in which different embodiments may be implemented. Many modifications to the described environment may be made by those skilled in the art without departing from the scope of the invention as described in the claims.

[0052] In step 201, the downsampling component 140 receives and / or retrieves DNA sequence data. In various embodiments of the invention, the downsampling component 140 may receive and / or retrieve DNA sequence data from a database, or as input from a user or another program.

[0053] In step 202, the downsampling component 140 performs bootstrapping on the DNA sequence data. In various embodiments of the invention, the downsampling component 140 performs bootstrapping on the received and / or retrieved DNA sequence data via the encoder 142. In various embodiments of the invention, the downsampling component 140 uses a modified variational encoder (e.g., encoder 142) to perform probabilistic resampling to obtain a resampled set including a plurality of reads, such as Figure 2BAs shown. The selected resampling can be self-consistent and representative of the original file, where self-consistency describes the minimum change that occurs between resamplings of the same size for a given number of resampling iterations, and representativeness is the minimization of the resampling changes relative to the original file. In various embodiments of the present invention, the downsampling component 140 uses equations 2-5 (described above) to ensure that its genomic integrity is maintained. In order to obtain a set of self-consistent resamplings, the downsampling component 140 compares similarities in the sampling distribution, and in doing so, it uses a hash table in the encoder to perform distribution calculations.

[0054] For example, Figure 2B An example of an overview of the bootstrapping process is provided. Figure 2B In the equation, θ k Sample 1 of (X1) undergoes random sampling 250 and generates a number 'n' of bootstrap samples, where 'n' is a predetermined number. In this particular example, the encoder 142 output is represented as θ k (X 11 ) of the bootstrap sampling 251, denoted as θ k (X 12 ) of the bootstrap sample 252, and denoted as θ k (X 1NR ) bootstrap sampling 253.

[0055] In step 204, the downsampling component 140 generates the biological information. In various embodiments of the present invention, the downsampling component 140 generates Figure 2C The biological information may include position information, genotype likelihood, and / or any other biological information known in the art.

[0056] In step 206, the downsampling component 140 generates a file with position and genotype information. In various embodiments of the present invention, the downsampling component 140 generates a file with position and genotype information. Figure 2C The position and genotype information is further described in the .

[0057] In step 208, the downsampling component 140 integrates the mapping position information and the genotype likelihood. In various embodiments, the downsampling component 140 can integrate the mapping position information and the genotype likelihood via the analysis engine 146 to identify the optimal vector representation of the resampling. The analysis engine 146 calculates the interval distribution of the multiple reads in each resampling so that the nucleotide position coverage is maximized, and finds the multiple reads with the minimum overlap in the resampling, and the minimum overlap has the maximum nucleotide coverage distribution of the given sampled genome. The analysis engine 146 can also calculate the average read weight based on the total weight of the resampling of the multiple genotype likelihoods occurring on each read and the weights of the multiple reads.

[0058] In step 210, downsampling component 140 may identify the best Z. In various embodiments of the invention, downsampling component 140 may identify the best Z via analysis engine 146. In various embodiments, downsampling component 140 may compute a list of best Zs via analysis engine 146 and display it to a user.

[0059] In step 212, the downsampling component 140 decodes the optimal resampled vector representation. In various embodiments of the invention, the downsampling component 140 can decode the optimal resampled vector representation via the decoder 144 to obtain a downsampled read file that is similar to and maintains the genomic integrity of the original file.

[0060] Figure 2C The diagram depicts Figure 1 A block diagram of a downsampling component 140 in communication with a computing device 110 and / or a server computer 120 within a distributed data processing environment 100 of an embodiment of the present invention is provided, wherein the downsampling component 140 is used to downsample DNA sequence data using a variational autoencoder. Figure 2C An overview of bioinformatic generation is provided, in which read fragments are mapped and their positional information and genotype likelihood are encoded into numerical representations. Figure 2C An illustration of an implementation is provided and does not imply any limitation on the environments in which different embodiments may be implemented. Many modifications to the described environment may be made by those skilled in the art without departing from the scope of the invention as described in the claims.

[0061] In step 204, the downsampling component 140 generates biological information. In various embodiments of the invention, the downsampling component 140 enables resampling of DNA fragments generated from massively parallel sequencing without compromising the genomic integrity of the original files.

[0062] In step 204A, the downsampling component 140 aligns the nucleotides. In various embodiments of the invention, the downsampling component 140 utilizes a modified variational encoder (eg, encoder 142), wherein the encoder 142 aligns multiple reads comprising nucleotides to different positions on the sampled genome.

[0063] In step 204B, the downsampling component 140 generates the position information of the aligned multiple readings. In various embodiments of the present invention, the downsampling component 140 can generate and / or identify the position information of the nucleotides in the aligned multiple readings via the encoder 142, wherein the position information includes: the mapping position of the mapped multiple readings, the mapping score and the nucleotide quality score. In various embodiments of the present invention, the downsampling component 140 can encode the mapping position, the mapping score and the nucleotide quality score of the multiple readings mapped to the specific position of the sampled genome into a vector representation via the encoder 142. For example, as known in the art, the FASTQ file with millions of readings is positioned to the reference genome using the Smith-Waterman algorithm. The Smith-Waterman algorithm performs local sequence alignment to determine the similar region between two nucleic acid sequences. The mapping position of each reading is then encoded into a matrix as the start and end position (matrix 220). In the matrix 220, the matrix structure includes the start and end positions of the readings with a data structure [n1, and n2] interval continuously, until biological data and / or multiple readings are represented, wherein n can be any position, integer and / or numerical representation known in the art.

[0064] In step 204C, the downsampling component 140 obtains the genotype likelihood of one or more genomic positions from the multiple reads, wherein the genotype likelihood is a plurality of alleles having a specified position variation on the multiple reads. In various embodiments of the invention, the encoder 142 is modified to obtain the genotype likelihood of each genomic position from the multiple reads using a Poisson graphical distribution model (Equation 1). In various embodiments of the invention, the downsampling component 140 determines, by the encoder 142, that the genotype likelihood matches the multiple reads, wherein the genotype likelihood gives the probability of the allele appearing at the nucleotide position, and discards the low genotype likelihood probability that is not supported by the multiple sequence reads.

[0065] In various embodiments of the present invention, a modified encoder (e.g., encoder 142) uses multiple genotype likelihoods on the reads to calculate and determine the read weights, reduces the weights of the reads with a few genotype likelihoods, and discards sequences that do not have any genotype likelihoods estimated at any nucleotide position and sequences that are not mapped to any position on the sampled genome to minimize redundancy. In various embodiments of the present invention, a modified encoder (e.g., encoder 142) can be used to construct representations from genotype likelihoods. The representation captures the genomic nuances in the original file so that the genomic variation in the original file is maintained in the downsampled representation.

[0066] In order to calculate the genotype likelihood on each read, the position of the reference genome is not considered independently, but the accumulation of genotypes is considered at the read level in combination. This enables positional read anomalies to be considered as dependent rather than assuming independence. Assuming that the occurrence of overlapping positions of reads with genotype accumulation is random, PGD can be used to model genotype likelihood. PGD takes into account the correlation of reads with positions and pulls the counts of reads together.

[0067] In addition, PGD regards the reading relative to the reference overlapping genomic position as the exponential family of multinomial joint distribution.This makes it possible to model the dependence between the counts of the readings overlapping with the position assuming the upper limit, which is reasonable for considering the depth of reading in particular.The likelihood genotype call can be estimated as the probability of the change from the reference allele to the alternative allele by the joint likelihood function.The reading overlapping with the genotype position is encoded in a matrix (e.g., matrix 222), which is used for training the model for estimating the genotype likelihood probability.The estimated genotype likelihood probability is also used for weighted reading.Therefore, ideally, most of the readings that have accumulated more genotype likelihoods can be found in the same resampling, thereby ensuring that the genome integrity of the original file is maintained.The resampling with the average cumulative genotype likelihood is weighted upward to increase its selection opportunity.

[0068] Figure 3 The embodiment according to the present invention is shown in Figure 1 Operational steps of a downsampling component 140 (generally indicated as 300) in communication with a computing device 110 within a distributed data processing environment 100 of the present invention, wherein the downsampling component 140 is used to automatically downsample DNA sequence data using a variational autoencoder. Figure 3 An illustration of an implementation is provided and does not imply any limitation on the environments in which different embodiments may be implemented. Many modifications to the described environment may be made by those skilled in the art without departing from the scope of the invention as described in the claims.

[0069] In step 302, the downsampling component 140 retrieves genomic sequence data from a file. In various embodiments of the invention, the downsampling component 140 may retrieve and / or receive genomic sequence data from one or more files, one or more databases, one or more search engines, and / or manually input data.

[0070] In step 304, the downsampling component 140 performs bootstrapping to generate resamples. In various embodiments of the invention, the downsampling component 140 may perform bootstrapping on the received and / or retrieved genomic sequence data to generate resamples via the encoder 142. In various embodiments of the invention, the downsampling component 140 may perform probabilistic resampling via a modified variational encoder (e.g., encoder 142) to obtain a resampled set comprising a plurality of reads.

[0071] In step 306, the downsampling component 140 resamples the DNA fragments. In various embodiments of the present invention, the downsampling component 140 can resample the DNA fragments generated from the parallel sequencing without compromising the genomic integrity of the original files.

[0072] In step 308, the downsampling component 140 generates biological information. In various embodiments of the present invention, the downsampling component 140 can generate biological information through the encoder 142, wherein the read segments are mapped, and the position information and genotype likelihood of the read segments are encoded into digital representations, such as Figure 2C shown.

[0073] In step 310, the downsampling component 140 generates a file containing position and genotype information. In various embodiments of the invention, the downsampling component 140, via the encoder 142, can generate one or more files containing position and genotype information.

[0074] In step 312, the downsampling component 140 accesses the non-representativeness and self-inconsistency of the generated resamples. In various embodiments of the invention, the downsampling component 140 may access the non-representativeness and self-inconsistency of one or more of the generated resamples via the encoder 142 and select one or more representative resamples.

[0075] At step 314, the downsampling component 140 builds a representation from the genotype likelihoods. In various embodiments of the invention, the downsampling component 140, via an encoder 142 (eg, a modified encoder), can build one or more vector representations from the genotype likelihoods.

[0076] At step 316, the downsampling component 140 integrates the mapping position information and the genotype likelihood. In various embodiments of the invention, the downsampling component 140 can integrate the mapping position information and the genotype likelihood via the analysis engine 146 to identify one or more optimal vector representations of one or more resamples.

[0077] In step 318, the downsampling component 140 decodes the resampled vector representation. In various embodiments of the invention, the downsampling component 140 may receive as input the representative sampling via the decoder 144 and produce a reconstruction of the original input reads. In various embodiments of the invention, the downsampling component 140 decodes the optimal resampled vector representation via the decoder 144 (e.g., a modified decoder) to obtain a downsampled read file that is similar to and maintains the genomic integrity of the original file.

[0078] In step 320, the downsampling component 140 outputs a downsampled read file. In various embodiments of the invention, the downsampling component 140 can output one or more downsampled read files to one or more users via the decoder 144 by displaying the data on the UI 106 and / or displaying a response prompt to remind the user of the file data, which is similar to and maintains the genomic integrity of the original file. In various embodiments of the invention, the downsampling component 140 through the decoder 144 can execute one or more downsampled read files that are similar to and maintain the genomic integrity of the original file.

[0079] Figure 4 The present invention is described in detail. Figure 1 1 is a block diagram of components of a server computer 120 in a distributed data processing environment 100. It should be understood that Figure 4 This merely provides an illustration of one implementation and does not imply any limitation with respect to the environments in which different embodiments may be implemented.Many modifications to the described environments are possible.

[0080] Figure 4 A computer system 400 is depicted, wherein the server computer 120 represents an example of a computer system 400 that includes a downsampling component 140. The computer system includes a processor 401, a cache 403, a memory 402, a permanent storage device 405, a communication unit 407, (one or more) input / output (I / O) interface 406, a display 409, (one or more) external devices 408, and a communication structure 404. The communication structure 404 provides communication between the cache 403, the memory 402, the permanent storage device 405, the communication unit 407, and the input / output (I / O) interface 406. The communication structure 404 can be implemented with any architecture designed to transfer data and / or control information between a processor (such as a microprocessor, a communication and network processor, etc.), a system memory, peripheral devices, and any other hardware components within the system. For example, the communication structure 404 can be implemented with one or more buses or crossbar switches.

[0081] Memory 402 and permanent storage 405 are computer-readable storage media. In this embodiment, memory 402 includes random access memory (RAM). In general, memory 402 may include any suitable volatile or non-volatile computer-readable storage media. Cache 403 is a fast memory that enhances the performance of processor 401 by saving recently accessed data from memory 402 and data near recently accessed data.

[0082] Program instructions and data for implementing embodiments of the present invention may be stored in permanent storage 405 and memory 402 for execution by one or more corresponding processors 401 via cache 403. In one embodiment, permanent storage 405 includes a magnetic hard drive. As an alternative or in addition to a magnetic hard drive, permanent storage 405 may include a solid-state hard drive, a semiconductor memory device, a read-only memory (ROM), an erasable programmable read-only memory (EPROM), flash memory, or any other computer-readable storage medium capable of storing program instructions or digital information.

[0083] The media used by permanent storage 405 also may be removable. For example, a removable hard drive may be used for permanent storage 405. Other examples include optical and magnetic disks, thumb drives, and smart cards that are inserted into a drive for transfer to another computer-readable storage medium that is also part of permanent storage 405.

[0084] In these examples, communication unit 407 provides communication with other data processing systems or devices. In these examples, communication unit 407 includes one or more network interface cards. Communication unit 407 can provide communication by using one or both of physical and wireless communication links. Program instructions and data for implementing embodiments of the present invention can be downloaded to permanent storage 405 through communication unit 407.

[0085] (One or more) I / O interface 406 enables input and output of data with other devices that can be connected to each computer system. For example, I / O interface 406 can provide connection to external device 408, and the external device is such as keyboard, keypad, touch screen and / or some other suitable input device. External device 408 can also include portable computer readable storage medium, such as, for example, thumb drive, portable optical disk or disk, and memory card. Software and data for practicing embodiments of the present invention can be stored on such portable computer readable storage medium, and can be loaded into permanent storage device 405 via (one or more) I / O interface 406. (One or more) I / O interface 406 is also connected to display 409.

[0086] Display 409 provides a mechanism for displaying data to a user and may be, for example, a computer monitor.

[0087] The programs described herein are identified based on the applications in which they are implemented in specific embodiments of the invention. However, it should be understood that any specific program terminology herein is used only for convenience, and thus the present invention should not be limited to use only in any specific application identified and / or implied by such terminology.

[0088] The present invention may be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium(s) having computer-readable program instructions thereon for causing a processor to perform aspects of the present invention.

[0089] A computer-readable storage medium may be any tangible device that can retain and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of computer-readable storage media includes the following: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device such as a punch card or a raised structure in a groove on which instructions are recorded, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium should not be interpreted as a temporary signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagated by a waveguide or other transmission medium (e.g., a light pulse by an optical fiber cable), or an electrical signal transmitted by a wire.

[0090] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to a corresponding computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical transmission fibers, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions to be stored in a computer-readable storage medium within the corresponding computing / processing device.

[0091] The computer-readable program instructions for performing the operation of the present invention can be assembly instructions, instruction set architecture (ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​(such as Smalltalk, C++, etc.) and conventional procedural programming languages ​​(such as "C" programming language or similar programming languages). The computer-readable program instructions can be executed completely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or completely on a remote computer or server. In the latter case, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider through the Internet). In certain embodiments, in order to perform various aspects of the present invention, the electronic circuit including, for example, a programmable logic circuit, a field programmable gate array (FPGA) or a programmable logic array (PLA) can execute the computer-readable program instructions to personalize the electronic circuit by utilizing the state information of the computer-readable program instructions.

[0092] Various aspects of the present invention are described herein with reference to the flow chart and / or block diagram of the method, device (system) and computer program product according to embodiments of the present invention. It will be understood that each frame of the flow chart and / or block diagram and the combination of frames in the flow chart and / or block diagram can be implemented by computer-readable program instructions.

[0093] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device create a device for implementing the functions / actions specified in one or more boxes of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium, which can guide the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable storage medium having the instructions stored therein includes an article of manufacture, which includes instructions for implementing aspects of the functions / actions specified in one or more boxes of the flowchart and / or block diagram.

[0094] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device, so that a series of operational steps are performed on the computer, other programmable apparatus, or other device to produce a computer-implemented process, so that the instructions executed on the computer, other programmable apparatus, or other device implement the functions / actions specified in one or more boxes of the flowchart and / or block diagram.

[0095] The flow chart and block diagram in the figure show the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention.In this regard, each frame in the flow chart or block diagram can represent a module, segment or part of an instruction, which includes one or more executable instructions for realizing a specified logical function.In some alternative embodiments, the function marked in the frame may not occur in the order marked in the figure, for example, the two frames shown in succession can actually be performed substantially at the same time, or these frames can sometimes be performed in reverse order, depending on the function involved.It will also be noted that each frame of the block diagram and / or flow chart illustration and the combination of frames in the block diagram and / or flow chart illustration can be realized by a dedicated hardware-based system that performs a specified function or action or performs a combination of special hardware and computer instructions.

[0096] The description of various embodiments of the present invention has been given for the purpose of illustration, but it is not intended to be exhaustive or limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the present invention. The terms used herein are selected to best explain the principles of the embodiments, practical applications, or technical improvements existing in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A computer-implemented method for automatically downsampling DNA sequence data using a variational autoencoder while maintaining the genomic integrity of the original file, the method comprising: processing the raw file by tagging one or more reads with a unique identifier and mapping the reads to a sampled reference sequence; performing, by an encoder, bootstrapping the genomic sequence data to generate resamples based on an original file, wherein the encoder selects a subset of reads from the original file using a probabilistic resampling framework and performs optimization of the bootstrapping to generate a resampled set for a certain size; evaluating, by the encoder, representativeness and self-consistency of the resamples, and selecting a representative resample based on the evaluation, wherein self-consistency describes the minimum variation that occurs between resamples of the same size for a given number of resample iterations, and representativeness is the minimization of the variation of the resamples relative to the original file; constructing, by the encoder, a vector representation from genotype likelihoods based on the selected representative resampling; Mapping position information and genotype likelihood are integrated by the analysis engine to identify the best vector representation for resampling; as well as The resampled identified optimal vector representation is decoded by a decoder to obtain a downsampled read file that is similar to the original file and maintains the genomic integrity of the original file, wherein the decoder produces a final set of read fragments representing the original file, and wherein the decoder reconstructs vector representations that encode the reads as they move through the pipeline to generate the downsampled read file.

2. The computer-implemented method of claim 1 , further comprising: The encoder resamples DNA fragments generated from massively parallel sequencing without compromising the genomic integrity of the original files.

3. The computer-implemented method of claim 1 , further comprising: Probabilistic resampling is performed by the encoder to obtain a resampled set comprising a plurality of readings.

4. The computer-implemented method of claim 1 , further comprising: receiving, by the decoder, the representative resample as input; as well as A reconstruction of the plurality of original input readings is produced by the decoder based on the received representative resamples.

5. The computer-implemented method of claim 1 , further comprising: A downsampled read file is output to the user that is similar to the original file and maintains the genomic integrity of the original file.

6. The computer-implemented method of claim 1 , further comprising: A file containing position and genotype information is generated by the encoder.

7. The computer-implemented method of claim 1, further comprising: The encoder generates biological information, wherein generating the biological information comprises: A plurality of reads containing nucleotides are aligned to different positions on the sampled genome by a modified variational encoder; generating, by the encoder, position information of the nucleotides in the aligned plurality of reads, wherein the position information comprises: a mapping position, a mapping score, and a nucleotide quality score of the mapped plurality of reads; encoding, by the encoder, the mapping positions, the mapping scores, and the nucleotide quality scores of a plurality of reads mapped to specific positions of a sampled genome into a vector representation; as well as Genotype likelihoods for one or more genomic positions are obtained by the encoder from the plurality of reads, wherein the genotype likelihoods are a plurality of alleles having a variation at a specified position across the plurality of reads.

8. A computer program product for automatically downsampling DNA sequence data using a variational autoencoder and maintaining the genomic integrity of the original file, the computer program product comprising: One or more computer-readable storage devices and program instructions stored on the one or more computer-readable storage devices, the stored program instructions comprising: program instructions for processing the raw file by tagging one or more reads with a unique identifier and mapping the reads to a sampled reference sequence; Instructions for a program to perform bootstrapping on genomic sequence data by an encoder to generate resamples based on the original file Let, wherein the encoder selects a subset of readings from the original file using a probabilistic resampling framework and performs a bootstrapped optimization to generate a resampled set for a certain size; program instructions for evaluating, by the encoder, representativeness and self-consistency of the resamples and selecting a representative resample based on the evaluation, wherein self-consistency describes the minimum variation that occurs between resamples of the same size for a given number of resample iterations and representativeness is the minimization of variation of the resample relative to the original file; program instructions for constructing, by the encoder, a vector representation from genotype likelihoods based on the selected representative resampling; Program instructions for integrating, by an analysis engine, the mapping position information and the genotype likelihoods to identify an optimal vector representation for resampling; and Program instructions for decoding, by a decoder, the resampled identified best vector representation to obtain a downsampled read file that is similar to the original file and maintains the genomic integrity of the original file, wherein the decoder produces a final set of read fragments representing the original file, and wherein the decoder reconstructs vector representations that encode reads as they move through the pipeline to generate the downsampled read file.

9. The computer program product of claim 8, further comprising: Program instructions for resampling, by the encoder, DNA fragments generated from massively parallel sequencing without compromising the genomic integrity of the original files.

10. The computer program product of claim 8, further comprising: Program instructions for performing probabilistic resampling by the encoder to obtain a resampled set comprising a plurality of readings.

11. The computer program product of claim 8, further comprising: program instructions for receiving, by the decoder, the representative resample as input; as well as Program instructions for producing, by the decoder, a reconstruction of a plurality of original input readings based on the received representative resamples.

12. The computer program product of claim 8, further comprising: Program instructions for outputting to a user the downsampled read file that is similar to the original file and maintains the genomic integrity of the original file.

13. The computer program product of claim 8, further comprising: Program instructions for generating, by the encoder, a file containing position and genotype information.

14. The computer program product of claim 8, further comprising: Program instructions for generating, by the encoder, biological information, wherein generating the biological information comprises: program instructions for aligning a plurality of reads containing nucleotides to different locations on a sampled genome by a modified variational encoder; Program instructions for generating, by the encoder, position information for nucleotides in the aligned plurality of reads, wherein the position information comprises: a mapping position, a mapping score, and a nucleotide quality score for the mapped plurality of reads; program instructions for encoding, by the encoder, mapping positions, mapping scores, and nucleotide quality scores of a plurality of reads mapped to specific positions of a sampled genome into a vector representation; and Program instructions for obtaining, by the encoder, genotype likelihoods for one or more genomic positions from the plurality of reads, wherein the genotype likelihoods are a plurality of alleles having a specified position variation across the plurality of reads.

15. A computer system for automatically downsampling DNA sequence data using a variational autoencoder and maintaining the genomic integrity of the original file, the computer system comprising: one or more computer processors; one or more computer-readable storage devices; Program instructions stored on the one or more computer-readable storage devices for execution by at least one of the one or more computer processors, the stored program instructions comprising: program instructions for processing the raw file by tagging one or more reads with a unique identifier and mapping the reads to a sampled reference sequence; Instructions for a program to perform bootstrapping on genomic sequence data by an encoder to generate resamples based on the original file Let, wherein the encoder selects a subset of readings from the original file using a probabilistic resampling framework and performs a bootstrapped optimization to generate a resampled set for a certain size; program instructions for evaluating, by the encoder, representativeness and self-consistency of the resamples and selecting a representative resample based on the evaluation, wherein self-consistency describes the minimum variation that occurs between resamples of the same size for a given number of resample iterations and representativeness is the minimization of variation of the resample relative to the original file; program instructions for constructing, by the encoder, a vector representation from genotype likelihoods based on the selected representative resampling; Program instructions for integrating, by an analysis engine, the mapping position information and the genotype likelihoods to identify an optimal vector representation for resampling; and Program instructions for decoding, by a decoder, the resampled identified best vector representation to obtain a downsampled read file that is similar to the original file and maintains the genomic integrity of the original file, wherein the decoder produces a final set of read fragments representing the original file, and wherein the decoder reconstructs vector representations that encode reads as they move through the pipeline to generate the downsampled read file.

16. The computer system of claim 15, further comprising: Program instructions for resampling, by the encoder, DNA fragments generated from massively parallel sequencing without compromising the genomic integrity of the original files.

17. The computer system of claim 15, further comprising: Program instructions for performing probabilistic resampling by the encoder to obtain a resampled set comprising a plurality of readings.

18. The computer system of claim 15, further comprising: program instructions for receiving, by the decoder, the representative resample as input; as well as Program instructions for producing, by the decoder, a reconstruction of a plurality of original input readings based on the received representative resamples.

19. The computer system of claim 15, further comprising: program instructions for outputting to a user the downsampled read file that is similar to the original file and maintains the genomic integrity of the original file; as well as Program instructions for generating, by the encoder, a file containing position and genotype information.

20. The computer system of claim 15, further comprising: Program instructions for generating, by the encoder, biological information, wherein generating the biological information comprises: program instructions for aligning a plurality of reads containing nucleotides to different locations on a sampled genome by a modified variational encoder; Program instructions for generating, by the encoder, position information for nucleotides in the aligned plurality of reads, wherein the position information comprises: a mapping position, a mapping score, and a nucleotide quality score for the mapped plurality of reads; program instructions for encoding, by the encoder, mapping positions, mapping scores, and nucleotide quality scores of a plurality of reads mapped to specific positions of a sampled genome into a vector representation; and Program instructions for obtaining, by the encoder, genotype likelihoods for one or more genomic positions from the plurality of reads, wherein the genotype likelihoods are a plurality of alleles having a specified position variation across the plurality of reads.

Images