{"id":33120239,"url":"https://github.com/bihealth/spliceai-wrapper","last_synced_at":"2025-11-19T22:01:19.915Z","repository":{"id":46083076,"uuid":"199652680","full_name":"bihealth/spliceai-wrapper","owner":"bihealth","description":"Wrapper for Illumina SpliceAI that caches results","archived":true,"fork":false,"pushed_at":"2022-12-26T20:39:56.000Z","size":44,"stargazers_count":3,"open_issues_count":3,"forks_count":1,"subscribers_count":6,"default_branch":"master","last_synced_at":"2025-09-09T16:10:39.710Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/bihealth.png","metadata":{"files":{"readme":"README.rst","changelog":"HISTORY.rst","contributing":"CONTRIBUTING.rst","funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2019-07-30T13:01:29.000Z","updated_at":"2023-10-17T10:26:32.000Z","dependencies_parsed_at":"2023-01-31T01:31:05.705Z","dependency_job_id":null,"html_url":"https://github.com/bihealth/spliceai-wrapper","commit_stats":null,"previous_names":[],"tags_count":1,"template":false,"template_full_name":null,"purl":"pkg:github/bihealth/spliceai-wrapper","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bihealth%2Fspliceai-wrapper","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bihealth%2Fspliceai-wrapper/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bihealth%2Fspliceai-wrapper/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bihealth%2Fspliceai-wrapper/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/bihealth","download_url":"https://codeload.github.com/bihealth/spliceai-wrapper/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bihealth%2Fspliceai-wrapper/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":285335552,"owners_count":27154282,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-11-19T02:00:05.673Z","response_time":65,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"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":[],"created_at":"2025-11-15T04:00:41.807Z","updated_at":"2025-11-19T22:01:19.909Z","avatar_url":"https://github.com/bihealth.png","language":"Python","funding_links":[],"categories":["Variant calling \u0026 alternative splicing tools"],"sub_categories":["Learning tools"],"readme":"================\nSpliceAI Wrapper\n================\n\n.. image:: https://img.shields.io/pypi/v/spliceai-wrapper.svg\n        :target: https://pypi.python.org/pypi/spliceai-wrapper\n\n.. image:: https://img.shields.io/travis/bihealth/spliceai-wrapper.svg\n        :target: https://travis-ci.org/bihealth/spliceai-wrapper\n\n`Illumina SpliceAI \u003chttps://github.com/Illumina/SpliceAI\u003e`_ is a nice method for predicting the impact of variants on splicing.\nHowever, it is computationally very expensive (45k variants/hour on a GPU, a few hundred variants per hour and CPU core).\n\nThis project, **SpliceAI Wrapper**, is an attempt to use caching for reducing the number of required predictions.\nPlease note that the authors of SpliceAI Wrapper are unrelated to the authors of SpliceAI.\n\n------------\nInstallation\n------------\n\nI recommend to use Bioconda\n\n.. code-block:: bash\n\n    $ conda install spliceai-wrapper\n\nIf you're not installing from Bioconda, make sure that you have ``bcftools`` and ``spliceai`` installed and the executables in your path.\n\n----------------------------\nImporting Precomputed Scores\n----------------------------\n\nFirst, obtain the precomputed scores from the SpliceAI project (I'm using the genome-wide ones filtered to a score \u003e= 0.1 for space usage reasons).\nThen:\n\n.. code-block:: bash\n\n    $ spliceai-wrapper prepare \\\n        --release GRCh37 \\\n        --precomputed-db-path path/to/precomputed.sqlite3 \\\n        --precomputed-vcf-path path/to/whole_genome_filtered_spliceai_scores.vcf.gz\n\nThis will import the precomputed scores into a SQLite3 database.\nOn my workstation, it takes about 20 minutes.\n\n------------------------\nRunning SpliceAI Wrapper\n------------------------\n\nObtain the gene list text file from the SpliceAI project.\nThen:\n\n.. code-block:: bash\n\n    $ spliceai-wrapper annotate \\\n        --input-vcf INPUT.vcf.gz \\\n        --output-vcf OUTPUT.vcf.gz \\\n        --genes-tsv path/to/grch37.txt \\\n        --precomputed-db-path path/to/precomputed.sqlite3 \\\n        --cache-db-path path/to/cache.sqlite3 \\\n        --path-reference path/to/hs37d5.fa \\\n        --release GRCh37\n\nFor trying it out use the ``--head 500`` parameter.\n\nThis will first go through ``INPUT.vcf.gz`` and try to find precomputed or cached values for all variants.\nThese precomputed/cached values will be used for annotation.\nVariants that lie outside the genes defined in ``grch37.txt`` are ignored.\nSNVs that lie within the genes defined in ``grch37.txt`` and that are not precomputed will be ignored as well (it is assumed their score is \u003c0.1 otherwise they would appear).\n\nThe remaining variants will be written to a temporary VCF file and ``spliceai`` will be called on them.\nThe annotations from the output of ``spliceai`` will be cached and the output VCF file and VCF file with cache hits will be merged into ``OUTPUT.vcf.gz``.\n\nNotes:\n\n- The precomputation database is opened read-only so you also don't need write permissions to this file.\n- The cache file must be writeable by your user, of course.\n- The extension of your output file determines what format is used for writing it.\n  ``.bcf`` files are written as compressed BCF, ``.vcf.gz`` and ``.vcf.bgz`` are written as bgzip-ed VCF, all other files will be written as text VCF files.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbihealth%2Fspliceai-wrapper","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fbihealth%2Fspliceai-wrapper","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbihealth%2Fspliceai-wrapper/lists"}