[{"_1":2,"_175":-5,"_176":-5},"loaderData",{"_3":4,"_17":18,"_25":26,"_150":151},"root",{"_5":-5,"_6":-7,"_7":8,"_9":-7,"_10":11,"_12":-7,"_13":14,"_15":16},"currentUser","primaryEmail","siteVerification","6LeIxAcTAAAAAJcZVRqyHh71UMIEGNQ_MXjiZKhI","toast","enableDev",false,"incidentStatus","theme","system","contrast","normal","routes/_layout",{"_19":-5,"_20":21,"_22":21,"_23":11,"_24":21},"username","isAskGordonEnabled",true,"isUsingNewTemplate","showMinimalLanding","showHoneypotTrap","routes/_layout.r.$namespace.$reponame",{"_27":28,"_29":-5,"_30":11,"_31":11,"_32":33,"_34":35,"_36":-5,"_37":11,"_38":39,"_88":-5,"_89":90,"_139":-5,"_140":11,"_141":21,"_142":11,"_143":-5,"_144":11,"_145":-5,"_146":147,"_148":91,"_149":-7},"canonicalUrl","https://hub.docker.com/r/pkrusche/paragraph/tags","mcpServer","canSeeDockerfile","canSeeAutobuilds","namespace","pkrusche","reponame","paragraph","cloudBuildSource","isAutobuildsAvailable","repo",{"_40":41,"_42":43,"_45":46,"_47":48,"_49":11,"_50":51,"_52":53,"_54":55,"_57":35,"_32":33,"_58":59,"_60":44,"_61":62,"_63":64,"_65":66,"_67":-5,"_68":62,"_69":70,"_71":11,"_72":33,"_73":74,"_79":21,"_80":81,"_85":33,"_86":87},"categories",[],"content_types",[44],"image","date_registered","2018-09-03T09:56:23.383776Z","description","Paragraph builds","is_private","last_updated","2018-09-04T14:24:11.333271Z","last_modified","2024-10-16T13:48:34.145251Z","media_types",[56],"application/vnd.docker.container.image.v1+json","name","pull_count",1002,"repository_type","star_count",0,"status",1,"status_description","active","source","collaborator_count","full_description","# ParaGRAPH: a suite of graph-based genotyping tools\n\n\n* [Introduction](#Introduction)\n* [Quick Start](#QuickStart)\n* [System Requirements](#SystemRequirements)\n\t* [Hardware](#Hardware)\n\t* [Operating systems](#Operatingsystems)\n\t* [Helper scripts](#Helperscripts)\n\t* [Library Dependencies](#LibraryDependencies)\n* [Installation](#Installation)\n* [Using helper scripts to run paraGRAPH](#UsinghelperscriptstorunparaGRAPH)\n* [Using vcf2paragraph.py to run paraGRAPH](#Usingvcf2paragraph.pytorunparaGRAPH)\n* [Further Information](#FurtherInformation)\n\t* [Documentation](#Documentation)\n\t* [Links](#Links)\n* [License](#License)\n\n\n\n\n## Introduction\n\nAccurate genotyping of known variants is a critical component of clinical-grade\npipelines for analysis of whole-genome sequencing data. It enables us to\n\n- test a sample for presence of specific pathogenic variants,\n- evaluate newly discovered variants,\n- compute background genotype distributions in different populations.\n\nParaGRAPH aims to facilitate these tasks by providing an accurate genotyper\nfor known deletions, insertions, and substitutions. It can be applied to either\na single sample or to a large cohort consisting of hundreds and even thousands\nof samples.\n\nIn addition to providing genotyping for deletion / insertion / substitution events,\nParaGRAPH provides a suite of graph-based tools to enable joint alignment and genotyping\nfor other event types. In a typical workflow, we can use these tools to\n\n * construct a graph that represents variant alleles,\n * align reads from one or many BAM files to the graph,\n * use read counts on nodes and edges to re-genotype the variant.\n\nIn future releases we will add support for genotyping more variant types using\nready-made workflows.\n\nParagraph is under active development -- if you have any questions or find\nbugs, please don't hesitate to contact us by creating Github issues for this\nproject!\n\n## Quick Start\n\nAfter [installing](#Installation) the tool, we can try and run a small test\ndataset we have included to see if everything is working as expected.\n\nLet's assume ParaGRAPH is installed as follows:\n\n```bash\n$ export PARAGRAPH=~/paragraph-install\n```\nThe directory will contain the following files (and a few more):\n```\n├── bin # binaries\n│ ├── multigrmpy.py # Graph genotyping wrappers\n│ ├── multiparagraph.py\n│ ├── ...\n│ ├── paragraph # graph aligner\n│ ├── grmpy # graph aligner and genotyper (single site)\n│ ├── idxdepth # BAM/CRAM depth estimator\n│ ├── kmerstats # kmer statistics tool for graphs\n│ ├── ...\n│ ├── findgrm.py # helpers and conversion scripts\n│ ├── compare_alignments.py\n│ ├── merge_paragraph_json.py\n│ ├── msa2vcf.py # convert multiple-alignment FASTA files to VCF\n│ ├── paragraph2dot.py # converts paragraph JSON graphs to DOT (graphviz)\n│ └── vcf2paragraph.py # converts VCF files to paragraph JSON\n├── lib # python libraries\n└── share # test and example data\n ├── analysis\n ├── examples\n ├── getting-started\n ├── paragraph_output_examples\n ├── schema\n └── test-data\n```\n\nFor most small-scale applications, the entry point is the `multigrmpy.py` script:\n\n```bash\n$ ${PARAGRAPH}/bin/multigrmpy.py\nusage: multigrmpy.py [-h] -i INPUT -o OUTPUT -m MANIFEST -r REFERENCE\n\nmultigrmpy.py: error: the following arguments are required: -i/--input, -o/--out, -m/--manifest, -r/--reference-sequence\n```\n\nThis script implements a workflow to genotype variants on one or more\nsamples. It minimally requires three inputs:\n\n* a reference FASTA file,\n* a candidate file of variants, which can be in JSON or VCF format,\n* a manifest / list of BAM files and their statistics.\n\nThe output is a directory, here we use `/tmp/paragraph-test` (before re-running with the same\noutput path you may want to delete this directory).\n\n```bash\n$ ${PARAGRAPH}/bin/multigrmpy.py \\\n -r ${PARAGRAPH}/share/test-data/genotyping_test_2/swaps.fa \\\n -i ${PARAGRAPH}/share/test-data/genotyping_test_2/swaps.vcf \\\n -m ${PARAGRAPH}/share/test-data/genotyping_test_2/samples.txt \\\n -o /tmp/paragraph-test\n$ tree /tmp/paragraph-test\n```\n\n```\n/tmp/paragraph-test\n├── grmpy.log\n├── genotypes.json.gz # genotypes and additional information for our one sample\n├── genotypes.vcf.gz # genotypes in VCF format (only when input was also VCF)\n└── variants.json.gz # full variant graphs, converted from swaps.vcf\n```\n\nThe first file to look at is the file `grmpy.log`:\n\n```\n[2018-05-30 19:08:01.057] [Genotyping] [11589850564992724888] [critical] Starting alignment for sample SWAPS (1/1)\n[2018-05-30 19:08:01.058] [Genotyping] [13762672856748146659] [critical] Starting alignment for sample SWAPS (1/1)\n[2018-05-30 19:08:01.058] [Genotyping] [14316483420521963862] [critical] Starting alignment for sample SWAPS (1/1)\n[2018-05-30 19:08:01.059] [Genotyping] [17003373142924278091] [critical] Starting alignment for sample SWAPS (1/1)\n[2018-05-30 19:08:01.059] [Genotyping] [9576328718556000728] [critical] Starting alignment for sample SWAPS (1/1)\n[2018-05-30 19:08:01.059] [Genotyping] [14181681436824356172] [critical] Starting alignment for sample SWAPS (1/1)\n[2018-05-30 19:08:01.060] [Genotyping] [4886869389785116662] [critical] Starting alignment for sample SWAPS (1/1)\n[2018-05-30 19:08:01.060] [Genotyping] [6672028834885690059] [critical] Starting alignment for sample SWAPS (1/1)\n[2018-05-30 19:08:09.431] [Genotyping] [11589850564992724888] [critical] Sample SWAPS: Alignment 1 / 3 finished\n[2018-05-30 19:08:09.569] [Genotyping] [13762672856748146659] [critical] Sample SWAPS: Alignment 2 / 3 finished\n[2018-05-30 19:08:09.758] [Genotyping] [14316483420521963862] [critical] Sample SWAPS: Alignment 3 / 3 finished\n[2018-05-30 19:08:09.760] [Genotyping] [6672028834885690059] [critical] Genotyping finished for graph 2 / 3\n[2018-05-30 19:08:09.760] [Genotyping] [11589850564992724888] [critical] Genotyping finished for graph 1 / 3\n[2018-05-30 19:08:09.760] [Genotyping] [14181681436824356172] [critical] Genotyping finished for graph 3 / 3\n```\n\nIf everything worked well, the genotypes JSON file will give the graph used\nfor each candidate variant as well as genotypes and Hardy-Weinberg p-values\nfor each sample.\n\nIn the example above, we have genotyped three events. The BAM files were constructed\nto give the following genotypes for each event:\n\n| **Event** | **Chromosome** | **Genotype** |\n|:---------:|:---------------:|:---------------------:|\n| swap1 | chrA | REF/REF (homref) |\n| swap2 | chrB | S1/S1 (homalt) |\n| swap1 | chrC | REF/S1 (heterozygous) |\n\nWe can extract these genotypes from the output file using Python script at bin/paragraph-to-csv.py\n\n```\nbin/paragraph-to-csv.py /tmp/paragraph-test/genotype.json.gz --genotype-only\n```\n\nThe output will be:\n\n```\n#FORMAT=GT\n#ID SWAPS\nchrA:1500-1509 REF/REF\nchrB:1500-1509 S1/S1\nchrC:1500-1699 REF/S1\n```\n\nThe output JSON file contains more information which can be used to link\nevents back to the original VCF file, genotype likelihoods, and also to get the\ngenotypes of the individual breakpoints.\n\nIf the input is a VCF file, then the output folder will contain an updated\nVCF file which allows us to quickly compare the original genotypes from the\ninput VCF, and those obtained by grmpy:\n\n```\nbcftools query -f '[%GT\\t%OLD_GT]\\n' /tmp/paragraph-test/genotypes.vcf.gz\n```\n```\n0/0 0/0\n1/1 1/1\n0/1 0/1\n```\n\nWe can see that the genotypes match in our use case, as expected.\n\nIn [doc/multiple-samples.md](doc/multiple-samples.md), we show how ParaGRAPH can be run on multiple samples with snakemake.\n[Doc/genotyping-parameters.md](doc/genotyping-parameters.md) contains\nadditional information on how to adjust parameters for genotyping and variant\nfiltering.\n\n## System Requirements\n\n### Hardware\n\nA standard workstation with at least 8GB of RAM should be sufficient for compilation and testing of the program. The\nsystem requirements for running the actual data analysis depend on the number of\ninput variants and samples. It typically takes up to a few seconds to genotype a single event in one sample. We provide\nwrapper scripts to parallelize this process. We have genotyped hundreds of thousands of variants in a single\nsample in around 2-3 hours using 40 CPU cores.\n\n### Operating systems\n\nParaGRAPH must be compiled with g++ version 4.9.x or later, or with a recent version of Clang.\nWe use the C++11 standard, any Posix compliant compiler supporting this standard\nshould be usable. We have tested using g++ and Clang on the following systems:\n\n- Ubuntu 16.04 and CentOS 5-7,\n- macOS 10.11+,\n- Windows is not supported currently.\n\n### Helper scripts\n\nAll helper, wrappers, and conversion scripts are implemented in Python 3.6 (see\n[graph-tools.md](doc/graph-tools.md)). The following third-party packages are required:\n\n- Pysam (MIT license; https://github.com/pysam-developers/pysam),\n- Intervaltree (Apache 2.0 license; https://pypi.python.org/pypi/intervaltree),\n- Jsonschema (MIT license; https://pypi.python.org/pypi/jsonschema).\n\nThe complete list of requrements can be found in [requirements.txt](requirements.txt).\n\n### Library Dependencies\n\n- [Boost libraries](http://www.boost.org) version >= 1.5 is required. Boost can be obtained from\n [http://www.boost.org](http://www.boost.org) and is available under the Boost license:\n [http://www.boost.org/users/license.html](http://www.boost.org/users/license.html).\n\n You may use your system Boost version, on Ubuntu, you can install the required versions\n of Boost as follows:\n ```bash\n sudo apt install libboost-dev libboost-iostreams-dev libboost-program-options-dev \\\n libboost-math-dev libboost-system-dev libboost-filesystem-dev\n ```\n\n Paragraph includes a copy of Boost 1.61 which can be built automatically during the\n configuration process. If you prefer to use this over your system's version you can\n use the `-DUSE_SYSTEM_BOOST=FALSE` cmake option.\n\n If you have a different Boost version you have precompiled, please follow the instructions below.\n For versions of Boost that are not installed system-wide, we prefer to statically link Boost libraries\n to Paragraph executables.\n This requires that static Boost libraries are generated which can be achived as follows:\n\n ```bash\n cd ~\n wget http://downloads.sourceforge.net/project/boost/boost/1.65.0/boost_1_65_0.tar.bz2\n tar xf boost_1_65_0.tar.bz2\n cd boost_1_65_0\n ./bootstrap.sh\n ./b2 --prefix=$HOME/boost_1_65_0_install link=static install\n ```\n\n To point Cmake to your version of Boost use the `BOOST_ROOT` environment variable:\n\n ```bash\n export BOOST_ROOT=$HOME/boost_1_65_0_install\n # Now run cmake + build -- note you may have to erase your cache + build folder\n # for the new settings to become active.\n ```\n\n- Recent versions of Google Test and Google Mock are required.\n\n Copyright: 2008, Google Inc., see\n [https://github.com/google/googletest/blob/master/googletest/LICENSE](https://github.com/google/googletest/blob/master/googletest/LICENSE)\n and\n [https://github.com/google/googletest/blob/master/googlemock/LICENSE](https://github.com/google/googletest/blob/master/googlemock/LICENSE).\n\n This repository includes copy of Google Test/Mock is ([external/googletest-release-1.8.0.tar.gz](external/googletest-release-1.8.0.tar.gz)).\n It will be built automatically during the configuration stage of the Cmake build process.\n\n See also the LICENSE files contained in [external/googletest-release-1.8.0.tar.gz](external/googletest-release-1.8.0.tar.gz).\n\n- Htslib is required. We include a copy of Htslib from\n [https://github.com/samtools/htslib](https://github.com/samtools/htslib). Htslib is released under MIT license, copyright (C) 2012-2014 Genome Research Ltd.\n\n The version of htslib we use is picked up from [external/htslib.tar.gz](external/htslib.tar.gz) and will be deployed\n in the build folder in `scratch/htslib` (see [external/make_dependencies.sh](external/make_dependencies.sh)).\n\n- Spdlog is required and included in the source tree. Spdlog is a header-only logging library from https://github.com/gabime/spdlog (MIT license,\n Copyright (c) 2016 Gabi Melman., see\n [https://github.com/gabime/spdlog/blob/master/LICENSE](https://github.com/gabime/spdlog/blob/master/LICENSE)\n and the LICENSE file in [external/spdlog.tar.gz](external/spdlog.tar.gz)).\n\n The library will be deployed during the cmake configuration process. It can be found in the build folder in\n `scratch/spdlog` (see [external/make_dependencies.sh](external/make_dependencies.sh)).\n\n- A list of other included libraries can be found in the [LICENSE](LICENSE) file.\n\n## Installation\n\n- **Native Build**. First, checkout the repository like so:\n\n ```bash\n git clone https://github.com/Illumina/paragraph.git\n cd paragraph-tools\n ```\n\n Then create a new directory for the program and compile it there:\n\n ```bash\n # Create a separate build folder.\n cd ..\n mkdir paragraph-tools-build\n cd paragraph-tools-build\n\n # Configure\n # optional:\n # export BOOST_ROOT=\n cmake ../paragraph-tools\n\n # Make, use -j to use n parallel jobs to build, e.g. make -j4\n make\n ```\n\n- **Docker** We also provide a [Dockerfile](Dockerfile). To build a Docker image, run the following command inside the source\n checkout folder:\n\n ```bash\n docker build .\n ```\n\n Once the image is built you can find out its ID like this:\n\n ```bash\n docker images\n ```\n ```\n REPOSITORY TAG IMAGE ID CREATED VIRTUAL SIZE\n 259aa8c0c920 10 minutes ago 2.18 GB\n ```\n\n The default entry point is the `multigrmpy.py` script:\n\n ```bash\n sudo docker run -v `pwd`:/data 259aa8c0c920\n usage: multigrmpy.py [-h] -i INPUT -o OUTPUT -b BAM -r REF [-p READ_LEN]\n [-l MAX_REF_NODE_LEN]\n [--log LOG]\n ```\n\n The current directory can be accessed as `/data` inside the Docker container, see also\n [https://docs.docker.com/engine/reference/commandline/run/](https://docs.docker.com/engine/reference/commandline/run/).\n\n To override the default entrypoint run the following command to get an interactive shell in which the paragraph tools\n can be run:\n\n ```bash\n sudo docker run --entrypoint /bin/bash -it 259aa8c0c920\n ```\n\n## Using helper scripts to run paraGRAPH\n\nAfter this, you can run the `multigrmpy.py` script from the build/bin directory on an example dataset as follows:\n\n```bash\npython3 bin/multigrmpy.py -i share/test-data/round-trip-genotyping/candidates.json \\\n -m share/test-data/round-trip-genotyping/samples.txt \\\n -r share/test-data/round-trip-genotyping/dummy.fa \\\n -o test \\\n```\n\nThis runs a simple genotyping example for two test samples. The input files are as follows:\n\n* **candidates.json**: this specifies some candidate SV events.\n ```json\n [\n {\n \"ID\": \"test-ins\",\n \"chrom\": \"chr1\",\n \"start\": 161,\n \"end\": 162,\n \"ins\": \"TGGGGGG\"\n },\n {\n \"ID\": \"test-del\",\n \"chrom\": \"chr1\",\n \"start\": 161,\n \"end\": 163,\n \"ins\": \"\"\n }\n ]\n ```\n\nAlternatively, candidate SV events can be specified as vcf.\n* **candidates.vcf**: this specifies candidate SV events in a vcf format.\n ```\n ##fileformat=VCFv4.2\n ##FILTER=\n ##ALT=\n #CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO\n chr1\t161\ttest-ins\tT\tTGGGGGG\t.\t.\t.\n chr1\t161\ttest-del\tTC\tT\t.\t.\t.\n ```\n\n* **samples.txt**: Manifest that specifies some test BAM files. Required columns: ID, path, depth, read length. Optional column: sex. Tab delimited.\n ```\n id path depth read length sex\n sample1\tsample1.bam 1 50 male\n sample2\tsample2.bam 1 50 female\n sample3\tsample2.bam 1 50 unknown\n ```\n\n* **dummy.fa** a short dummy reference which only contains `chr1`\n\nThe output folder `test` then contains gzipped json for final genotypes:\n\n```bash\n$ tree test\n```\n```\ntest\n├── GraphTyping.log # main workflow log file\n├── variants.json # Json formatted candidate variant graphs. Only exist if input is a VCF.\n└── genotype.json.gz # genotyping result\n```\n\n## Using vcf2paragraph.py to run paraGRAPH\n\nConsider an insertion specified by the following VCF record:\n\n```\n#CHROM POS ID REF ALT QUAL FILTER INFO FORMAT ALT\nchr1 939570 . T TCCCTGGAGGACC 0 PASS . GT 1\n```\n\n![doc/breakpoint-genotyper.png](doc/breakpoint-genotyper.png)\n\nWe can construct a sequence graph that corresponds to this insertion (this uses\nthe `vcf2paragraph.py` and `paragraph2dot.py` scripts provided with this package,\nand a hg38 reference sequence which can be obtained from\n[http://hgdownload.cse.ucsc.edu/goldenPath/hg38/bigZips](http://hgdownload.cse.ucsc.edu/goldenPath/hg38/bigZips/)).\nThe `vcf2paragraph.py` script supports a subset of the VCF file format which is documented in\n[doc/graph-tools.md](doc/graph-tools.md#vcf2paragraph.py).\n\n```bash\npython3 vcf2paragraph.py share/test-data/paragraph/pg-het-ins/pg-het-ins.vcf pg-het-ins.json -r hg38.fa\n# visualize using dot / graphviz, see http://www.graphviz.org/\npython3 paragraph2dot.py pg-het-ins.json pg-het-ins.dot\ndot -Tpng pg-het-ins.dot > pg-het-ins.png\n```\n\nThe resulting graph has the following structure:\n\n![](doc/ins-graph.png)\n\nNodes and edges may be labelled using sequence tags such as `REF` and\n`ALT`. These induce paths in our sequence graph and identify the different alleles\nwe would like to genotype.\n\nThe JSON File [pg-het-ins.json](share/test-data/paragraph/pg-het-ins/pg-het-ins.json) contains\na description of the sequence graph derived from the input VCF file. We can align reads to this\ngraph using that `paragraph` tool as follows:\n\n```bash\nparagraph -r hg38.fa -g pg-het-ins.json -b share/test-data/paragraph/pg-het-ins/na12878.bam \\\n -o pg-het-ins-readcounts.json\n```\n\nThe resulting JSON file gives read counts and alignments (part of this figure was\ncreated using a modified version of SequenceTubeMap -- see https://github.com/vgteam/sequenceTubeMap):\n\n![](doc/ins-graph-figure.png)\n\nFor editing / manual inspection, it is typically a good idea to pretty-print the file:\n\n```bash\npython -mjson.tool pg-het-ins-readcounts.json > pg-het-ins-readcounts.pretty.json\n```\n\nThe resulting file contains the original graph in JSON format (see [pg-het-ins.json](share/test-data/paragraph/pg-het-ins/pg-het-ins.json))\n\n```javascript\n{\n \"sequencenames\": [ \"ALT\", \"REF\" ],\n \"nodes\": [\n {\n \"name\": \"source\",\n \"sequence\": \"NNNNNNNNNN\"\n },\n // ...\n ],\n \"edges\": [\n {\n \"from\": \"939571-939570:CCCTGGAGGACC\",\n \"to\": \"ref-chr1:939571-939719\"\n },\n // ...\n ],\n \"target_regions\": [ \"chr1:939420-939719\" ],\n```\n\nWe also have the paths induced by the edge labels (this was added by `vcf2paragraph.py`):\n\n```javascript\n \"paths\": [\n {\n \"nodes\": [ \"source\", \"ref-chr1:939420-939570\", \"939571-939570:CCCTGGAGGACC\", \"ref-chr1:939571-939719\", \"sink\" ],\n \"nucleotide_length\": 312,\n \"path_id\": \"ALT|1\",\n \"sequence\": \"ALT\"\n },\n {\n \"nodes\": [ \"source\", \"ref-chr1:939420-939570\", \"ref-chr1:939571-939719\", \"sink\" ],\n \"nucleotide_length\": 300,\n \"path_id\": \"REF|1\",\n \"sequence\": \"REF\"\n }\n ],\n```\n\nEach node, edge, and path has reads associated with it. We provide read counts for forward\nand reverse strands (`:READS`, `:FWD`, `:REV`) and fragment counts (these counts are corrected\nfor the same reads possibly originating from the same sequence fragment in the case of\npaired-end sequencing data).\n\n```javascript\n \"read_counts_by_edge\": {\n \"939571-939570:CCCTGGAGGACC_ref-chr1:939571-939719\": 15,\n \"939571-939570:CCCTGGAGGACC_ref-chr1:939571-939719:FWD\": 6,\n \"939571-939570:CCCTGGAGGACC_ref-chr1:939571-939719:READS\": 15,\n \"939571-939570:CCCTGGAGGACC_ref-chr1:939571-939719:REV\": 9,\n \"ref-chr1:939420-939570_939571-939570:CCCTGGAGGACC\": 14,\n // ...\n \"total\": 29,\n \"total:FWD\": 14,\n \"total:READS\": 29,\n \"total:REV\": 15\n },\n \"read_counts_by_node\": {\n \"939571-939570:CCCTGGAGGACC\": 14,\n \"939571-939570:CCCTGGAGGACC:FWD\": 6,\n \"939571-939570:CCCTGGAGGACC:READS\": 14,\n \"939571-939570:CCCTGGAGGACC:REV\": 8,\n \"ref-chr1:939420-939570\": 27,\n // ...\n },\n \"read_counts_by_sequence\": {\n \"ALT\": {\n \"total\": 14,\n \"total:FWD\": 6,\n \"total:READS\": 14,\n \"total:REV\": 8\n },\n \"REF\": {\n \"total\": 15,\n \"total:FWD\": 8,\n \"total:READS\": 15,\n \"total:REV\": 7\n }\n },\n```\n\nBased on the read count information provided by paragraph, we can further derive genotypes of each sample using the genotyper, grmpy. Further details about the genotyping method can be found in [graph-models.md](doc/graph-models.md).\n\nHere is one example about the genotyping result on a simple insertion with 3 nodes: left flank (LF), right flank (RF), insertion sequence (INS).\n\nIt is extracted and re-organized from [an expected output](share/test-data/multigrmpy/expected.json)):\n\n```javascript\n{\n // ...\n \"eventinfo\": { /* summary information about the event that was genotyped */},\n \"graphinfo\": { /* summary information about the graph that was created for that event */},\n \"samples\": { // genotyping information for all samples\n \"Dummy\": { // sample named \"Dummy\"\n \"Filter\": \"PASS\", // filter information for the variant genotyping\n \"GT\": \"REF/INS\", // most likely genotype of the variant\n \"num_support\": 2, // number of breakpoints supporting the most likely genotype\n \"otherGT\": { // additional information for other genotypes\n \"INS/INS\": {\n \"num_unsupport\": 2 // both breakpoint do not support this genotype\n },\n \"REF/REF\": {\n \"num_unsupport\": 2\n }\n },\n \"read_counts_by_edge\": { // number of reads aligned to each edge\n \"INS_RF\": 1,\n \"LF_INS\": 2,\n \"LF_RF\": 2\n }\n \"breakpoints\": { // detailed information for genotypes of each breakpoint\n \"LF_\": { // breakpoint starting at node \"LF\"\n \"GT\": \"REF/INS\", // most likely genotype of this breakpoint\n \"DP\": 4, // number of edge counts on this breakpoint\n \"GL\": { // genotype likelihood estimated from simple poisson model\n \"INS/INS\": -13.93799761671912,\n \"REF/INS\": -7.486945295275721,\n \"REF/REF\": -13.93799761671912\n },\n \"allele_fractions\": [ // fraction of edge counts for each allele\n 0.5,\n 0.5\n ]\n },\n \"_RF\": { // breakpoint ending at node \"RF\"\n \"GT\": \"REF/INS\",\n \"DP\": 3,\n \"GL\": {\n \"INS/INS\": -12.84913761949369,\n \"REF/INS\": -5.714988453343845,\n \"REF/REF\": -8.254017769359102\n },\n \"allele_fractions\": [\n 0.6666666666666666,\n 0.3333333333333333\n ]\n }\n }\n }\n }\n }\n // ...\n}\n```\n\n## Further Information\n\n### Documentation\n\n* More **information about all tools we provide in this package** can be found in\n [doc/graph-tools.md](doc/graph-tools.md).\n\n* In [doc/graph-models.md](doc/graph-models","has_starred","hub_user","immutable_tags_settings",{"_75":11,"_76":77},"enabled","rules",[78],".*","is_automated","permissions",{"_82":21,"_83":11,"_84":11},"read","write","admin","user","storage_size",1336967003,"orgData","recentTags",[91,129],{"_92":44,"_93":94,"_95":96,"_97":98,"_99":100,"_50":118,"_119":94,"_120":33,"_121":56,"_57":122,"_123":124,"_125":109,"_126":118,"_127":66,"_128":21},"content_type","creator",217140,"full_size",471315020,"id",34423051,"images",[101],{"_102":103,"_104":105,"_106":107,"_108":109,"_110":111,"_112":113,"_114":107,"_115":-5,"_116":96,"_63":66,"_117":-5},"architecture","amd64","digest","sha256:059b81676d1587e4af11e1f93f1022f238e90f2bd96dc60b50d4ce8837a15386","features","","last_pulled","2026-09-22T12:06:21.246247111Z","last_pushed","2018-09-03T11:18:38Z","os","linux","os_features","os_version","size","variant","2018-09-03T11:18:37.22228Z","last_updater","last_updater_username","media_type","latest","repository",5965456,"tag_last_pulled","tag_last_pushed","tag_status","v2",{"_92":44,"_93":94,"_95":130,"_97":131,"_99":132,"_50":137,"_119":94,"_120":33,"_121":56,"_57":138,"_123":124,"_125":135,"_126":137,"_127":66,"_128":21},865635330,34492743,[133],{"_102":103,"_104":134,"_106":107,"_108":135,"_110":136,"_112":113,"_114":107,"_115":-5,"_116":130,"_63":66,"_117":-5},"sha256:eeefde07e77e6036c876db0d7e863b2a5595cb463ed9ee1086181e0b4898eed2","2026-09-18T09:26:44.188179944Z","2018-09-04T14:24:11Z","2018-09-04T14:24:10.331496Z","ubuntu-1804","sbxKitMetadata","isPotentialSpam","isHubSbxKitsEnabled","showSbxLogoOverride","sbxTrademarkText","displaySecurityMetadata","tagAttestations","tagsHaveAttestations",[],"selectedTag","attestationsUrl","routes/_layout.r.$namespace.$reponame.tags",{"_38":152,"_159":160,"_146":173,"_174":11},{"_40":153,"_42":154,"_45":46,"_47":48,"_49":11,"_50":51,"_52":53,"_54":155,"_57":35,"_32":33,"_58":59,"_60":44,"_61":62,"_63":64,"_65":66,"_67":-5,"_68":62,"_69":70,"_71":11,"_72":33,"_73":156,"_79":21,"_80":158,"_85":33,"_86":87},[],[44],[56],{"_75":11,"_76":157},[78],{"_82":21,"_83":11,"_84":11},"tags",{"_161":162,"_163":-5,"_164":-5,"_165":166},"count",2,"next","previous","results",[167,170],{"_92":44,"_93":94,"_95":130,"_97":131,"_99":168,"_50":137,"_119":94,"_120":33,"_121":56,"_57":138,"_123":124,"_125":135,"_126":137,"_127":66,"_128":21},[169],{"_102":103,"_104":134,"_106":107,"_108":135,"_110":136,"_112":113,"_114":107,"_115":-5,"_116":130,"_63":66,"_117":-5},{"_92":44,"_93":94,"_95":96,"_97":98,"_99":171,"_50":118,"_119":94,"_120":33,"_121":56,"_57":122,"_123":124,"_125":109,"_126":118,"_127":66,"_128":21},[172],{"_102":103,"_104":105,"_106":107,"_108":109,"_110":111,"_112":113,"_114":107,"_115":-5,"_116":96,"_63":66,"_117":-5},[],"canDeleteTags","actionData","errors"]

pkrusche/paragraph

Tags for pkrusche/paragraph