Characterization of Germline variants
-
Updated
Mar 15, 2022 - Python
Characterization of Germline variants
Cancer Predisposition Sequencing Reporter (CPSR)
Bitscopic Interpreting ACMG Standards 2015
Tool for automated classification of genetic variants according to ACMG criteria.
ACMG Assistant is a student-level, research-oriented variant classification tool developed to explore the practical application of ACMG/AMP 2015 and 2023 guidelines. It combines automated retrieval of annotation data from public APIs with structured interactive evidence collection to support systematic variant interpretation.
The newest iteration of my personal web app.
Five Google ADK / Agent Builder agents watching genomic evidence (ClinVar, gnomAD, AlphaMissense) synced via the Fivetran MCP and, the moment a Variant of Uncertain Significance is reclassified, recompute a calibrated ACMG posterior to draft the patient recontact + family cascade no system sends today. Draft-only, FHIR R4, human-in-the-loop.
Research-only rare-disease evidence infrastructure, currently implemented and evaluated for TSC1/TSC2. Builds reproducible variant evidence packets, leakage-safe validation controls, and a source-grounded Mechanism Atlas. Not a clinical or diagnostic system; issues no authoritative variant classifications.
Calibrate functional/in-silico variant scores into ACMG clinical evidence strengths (ClinGen-SVI), with LDLR/AlphaMissense worked example
Drafts ACMG/AMP variant classifications for a human curator: gathers genomics evidence, adjudicates each criterion with Claude, computes the label in code via ClinGen points, and shows every source.
Narrows 2.3M ClinVar variants of uncertain significance to 306 cardiovascular reclassification candidates, combining gnomAD population frequencies with AlphaMissense predictions across 71.7M scored variants.
R pipeline for population-specific variant frequency analysis using gnomAD data — includes MAF calculation, Fisher's Exact Test with Bonferroni correction, and bubble plot visualization across 11 global populations
Hepatology rare-variant clinical intelligence — pulls 12 evidence sources in parallel, applies the ACMG/AMP classifier, and synthesizes with Claude Opus 4.7 + extended thinking. Built for the Cerebral Valley 'Built with Opus 4.7' hackathon (2026).
Feedback, bug reports, and feature requests for GeneBe — a genetic variant interpretation platform.
Agentic clinical variant interpretation copilot: drafts ACMG/AMP classifications with cited evidence, evaluated against ClinVar gold labels. RAG + LLM eval + LLMOps.
Machine-learning triage of LDLR variants of uncertain significance using predictor concordance and ACMG-aligned evidence mapping
Add a description, image, and links to the acmg topic page so that developers can more easily learn about it.
To associate your repository with the acmg topic, visit your repo's landing page and select "manage topics."