{"id":37133492,"url":"https://github.com/google/deepsomatic","last_synced_at":"2026-03-06T01:10:22.922Z","repository":{"id":203169296,"uuid":"703751361","full_name":"google/deepsomatic","owner":"google","description":"DeepSomatic is an analysis pipeline that uses a deep neural network to call somatic variants from tumor-normal and tumor-only sequencing data.","archived":false,"fork":false,"pushed_at":"2025-10-15T18:54:15.000Z","size":72,"stargazers_count":293,"open_issues_count":3,"forks_count":45,"subscribers_count":13,"default_branch":"r1.9","last_synced_at":"2026-02-22T22:21:04.240Z","etag":null,"topics":["bioinformatics","deep-learning","deepsomatic","dna","genome","genomics","machine","science","sequencing","somatic-mutations","somatic-variants"],"latest_commit_sha":null,"homepage":"","language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"bsd-3-clause","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/google.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2023-10-11T20:56:37.000Z","updated_at":"2026-02-21T18:47:15.000Z","dependencies_parsed_at":"2024-08-27T22:32:42.089Z","dependency_job_id":"1d6e1f3f-958a-4e3b-97e1-90dd6dcd8363","html_url":"https://github.com/google/deepsomatic","commit_stats":{"total_commits":10,"total_committers":1,"mean_commits":10.0,"dds":0.0,"last_synced_commit":"8810f640887b1875901d8ec93cdf3e3591e8c2e1"},"previous_names":["google/deepsomatic"],"tags_count":6,"template":false,"template_full_name":null,"purl":"pkg:github/google/deepsomatic","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/google%2Fdeepsomatic","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/google%2Fdeepsomatic/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/google%2Fdeepsomatic/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/google%2Fdeepsomatic/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/google","download_url":"https://codeload.github.com/google/deepsomatic/tar.gz/refs/heads/r1.9","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/google%2Fdeepsomatic/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":30157054,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-03-05T22:39:40.138Z","status":"ssl_error","status_checked_at":"2026-03-05T22:39:24.771Z","response_time":93,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.6:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["bioinformatics","deep-learning","deepsomatic","dna","genome","genomics","machine","science","sequencing","somatic-mutations","somatic-variants"],"created_at":"2026-01-14T15:35:15.811Z","updated_at":"2026-03-06T01:10:22.911Z","avatar_url":"https://github.com/google.png","language":null,"readme":"# DeepSomatic\n\n[![release](https://img.shields.io/badge/release-v1.9.0-green?logo=github)](https://github.com/google/deepvariant/releases)\n[![announcements](https://img.shields.io/badge/announcements-blue)](https://groups.google.com/d/forum/deepvariant-announcements)\n[![blog](https://img.shields.io/badge/blog-orange)](https://goo.gl/deepvariant)\n\nDeepSomatic is an extension of deep learning-based variant caller\n[DeepVariant](https://github.com/google/deepvariant) that takes aligned reads\n(in BAM or CRAM format) from tumor and normal data, produces pileup image\ntensors from them, classifies each tensor using a convolutional neural network,\nand finally reports somatic variants in a standard VCF or gVCF file.\n\nDeepSomatic supports somatic variant-calling from tumor-normal and tumor-only\nsequencing data.\n\n## Code availability\n\nDeepSomatic is integrated with\n[DeepVariant](https://github.com/google/deepvariant) to utilize the high-quality\nend-to-end testing and feature development of DeepVariant.\n\nHere are the scripts that describe the core components of DeepSomatic:\n\n*   [run_deepsomatic](https://github.com/google/deepvariant/blob/r1.9/scripts/run_deepsomatic.py):\n    The DeepSomatic runner script.\n\n*   [make_examples_somatic](https://github.com/google/deepvariant/blob/r1.9/deepvariant/make_examples_somatic.py):\n    The `make_examples` step for DeepSomatic.\n\n*   [call_variants](https://github.com/google/deepvariant/blob/r1.9/deepvariant/call_variants.py):\n    Inference script that generates the variant calls.\n\n*   [postprocess_variants](https://github.com/google/deepvariant/blob/r1.9/deepvariant/postprocess_variants.py):\n    Updated with `process_somatic` option to process somatic variants.\n\n*   [dockerfile](https://github.com/google/deepvariant/blob/r1.9/Dockerfile.deepsomatic):\n    The Dockerfile for DeepSomatic.\n\nIntegrating DeepSomatic within DeepVariant helps to maintain\nhigh-quality code health with integrated testing and feature development.\n\n## Case studies\n\nThe following case studies show example runs for supported technologies:\n\n### Tumor-normal case-studies\n\n*   Illumina WGS tumor-normal [case study](docs/deepsomatic-case-study-wgs.md).\n\n*   Illumina WES tumor-normal [case study](docs/deepsomatic-case-study-wes.md).\n\n*   PacBio tumor-normal [case study](docs/deepsomatic-case-study-pacbio.md).\n\n*   ONT tumor-normal [case study](docs/deepsomatic-case-study-ont.md).\n\n*   FFPE WGS tumor-normal [case study](docs/deepsomatic-case-study-ffpe-wgs.md).\n\n*   FFPE WES tumor-normal [case study](docs/deepsomatic-case-study-ffpe-wes.md).\n\n### Tumor-only case-studies\n\n*   Illumina WGS tumor-only\n    [case study](docs/deepsomatic-case-study-wgs-tumor-only.md).\n\n*   PacBio tumor-only\n    [case study](docs/deepsomatic-case-study-pacbio-tumor-only.md).\n\n*   ONT tumor-only [case study](docs/deepsomatic-case-study-ont-tumor-only.md).\n\n*   FFPE WGS tumor-only [case study](docs/deepsomatic-case-study-ffpe-wgs-tumor-only.md).\n\n*   FFPE WES tumor-only [case study](docs/deepsomatic-case-study-ffpe-wes-tumor-only.md).\n\nFor details around runtime and accuracy expectations, please see the [DeepSomatic metrics page](docs/metrics.md).\n\n## How to Cite\n\nIf you use DeepSomatic in your work, please cite:\n\n[DeepSomatic: Accurate somatic small variant discovery for multiple sequencing technologies]( https://doi.org/10.1101/2024.08.16.608331)\n\n## How to run DeepSomatic\n\n```bash\nsudo docker run \\\n-v ${INPUT_DIR}:${INPUT_DIR} \\\n-v ${OUTPUT_DIR}:${OUTPUT_DIR} \\\ngoogle/deepsomatic:\"${BIN_VERSION}\" \\\nrun_deepsomatic \\\n--model_type=WGS \\ ** Can be WGS,WES,PACBIO,ONT,FFPE_WGS,FFPE_WES,WGS_TUMOR_ONLY,PACBIO_TUMOR_ONLY,ONT_TUMOR_ONLY **\n--ref=${INPUT_DIR}/REF.fasta \\ **Path to reference fasta file.\n--reads_normal=${INPUT_DIR}/normal.bam \\ **Path to normal bam file.\n--reads_tumor=${INPUT_DIR}/tumor.bam \\ * Path to tumor bam file.\n--output_vcf=${OUTPUT_DIR}/OUTPUT.vcf.gz \\ **Path to output VCF file.\n--output_gvcf=${OUTPUT_DIR}/OUTPUT.g.vcf.gz \\ **Path to output gVCF file.\n--sample_name_tumor=\"tumor\" \\\n--sample_name_normal=\"normal\" \\\n--num_shards=$(nproc) \\ **Total number of threads to use.\n--logging_dir=${OUTPUT_DIR}/logs \\ **Log output directory.\n--intermediate_results_dir ${OUTPUT_DIR}/intermediate_results_dir \\\n--regions=chr1 \\ **Region of the genome, if not provided then runs on whole genome\n--use_default_pon_filtering=false \\ **Set to true for default PON filtering for tumor-only variant calling**\n--dry_run=false **Default is false. If set to true, commands will be printed out but not executed.\n```\n\nPlease follow the [Quick Start](docs/deepsomatic-quick-start.md) for more\ndetails on different setups like Docker and Singuarity. available for\nDeepSomatic\n\n### Example output\n\nDeepSomatic utilizes FILTER in VCF format to report identified germline and\nsomatic variants. The description of the filters can be found in the header:\n\n```bash\n##FILTER=\u003cID=PASS,Description=\"All filters passed\"\u003e\n##FILTER=\u003cID=RefCall,Description=\"Genotyping model thinks this site is reference.\"\u003e\n##FILTER=\u003cID=LowQual,Description=\"Confidence in this variant being real is below calling threshold.\"\u003e\n##FILTER=\u003cID=NoCall,Description=\"Site has depth=0 resulting in no call.\"\u003e\n##FILTER=\u003cID=GERMLINE,Description=\"Non somatic variants\"\u003e\n```\n\nFor example, the variants reported below:\n\n```bash\n# CHROM POS     ID  REF ALT QUAL    FILTER      INFO    FORMAT              SAMPLE_NAME\nchr1    14001   .   A   G   3.7     GERMLINE    .       GT:GQ:DP:AD:VAF:PL  0/0:4:8:4,4:0.5:1,0,34\nchr1    14002   .   T   A   0       RefCall     .       GT:GQ:DP:AD:VAF:PL  0/0:51:60:57,2:0.0333333:0,51,58\nchr1    14003   .   C   G   43.8    PASS        .       GT:GQ:DP:AD:VAF:PL  1/1:43:74:0,74:1:43,52,0\n```\n\nIn this example:\n\n* The variant with `GERMLINE` FILTER status is identified as a germline variant\n* The variant with `RefCall` FILTER status is homozygous to the reference\n* The variant with `PASS` FILTER status is a **somatic variant**.\n\n### Prerequisites\n\n*   Unix-like operating system (cannot run on Windows)\n*   Python 3.10\n\n## Contribution Guidelines\n\nPlease [open a pull request](https://github.com/google/deepsomatic/compare) if\nyou wish to contribute to DeepSomatic. Note, we have not set up the\ninfrastructure to merge pull requests externally. If you agree, we will test and\nsubmit the changes internally and mention your contributions in our\n[release notes](https://github.com/google/deepsomatic/releases). We apologize\nfor any inconvenience.\n\nIf you have any difficulty using DeepSomatic, feel free to\n[open an issue](https://github.com/google/deepsomatic/issues/new). If you have\ngeneral questions not specific to DeepSomatic, we recommend that you post on a\ncommunity discussion forum such as [BioStars](https://www.biostars.org/).\n\n## License\n\n[BSD-3-Clause license](LICENSE)\n\n## Disclaimer\n\nThis is not an official Google product.\n\nNOTE: the content of this research code repository (i) is not intended to be a\nmedical device; and (ii) is not intended for clinical use of any kind, including\nbut not limited to diagnosis or prognosis.\n","funding_links":[],"categories":[],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgoogle%2Fdeepsomatic","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fgoogle%2Fdeepsomatic","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgoogle%2Fdeepsomatic/lists"}