Skip to content
View Agni-18's full-sized avatar

Block or report Agni-18

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Agni-18/README.md

Agnidipa Sett

M.Tech Bioinformatics student at Delhi Technological University (graduating 2027). I build and validate bioinformatics pipelines for genomics, transcriptomics and drug discovery, and I test them against ground truth.

Projects

  • clingenomics: ACMG/AMP germline and AMP/ASCO/CAP somatic variant interpretation engine in Python, with 62 tests, CI and a ClinVar concordance harness.
  • ngs-variant-calling-pipeline: Snakemake and Docker pipeline benchmarking GATK against DeepVariant on NA12878 chr20 with the GIAB truth set (DeepVariant F1 99.02% for SNPs, 97.35% for indels).
  • rnaseq-analysis-suite: bulk, single-cell and long-read RNA-seq pipelines scored against simulated ground truth, with the real-data drop reported.
  • ad-target-dossier: harmonization of three Alzheimer's cohorts (494 samples) to ontology terms, with a Neo4j knowledge graph and Open Targets scores.
  • lung-adenocarcinoma-deg-analysis: differential expression in 107 lung adenocarcinoma samples and survival testing in 497 TCGA-LUAD patients.
  • g4-ligand-classifier: G-quadruplex ligand classifier with scaffold-grouped validation (CV AUC 0.94 on 367 compounds).

Tools

Python, R, SQL, Snakemake, Nextflow, Docker, GATK, DeepVariant, DESeq2, Scanpy, RDKit, scikit-learn, Neo4j

Pinned Loading

  1. ad-target-dossier ad-target-dossier Public

    Harmonizes 3 GEO Alzheimer's datasets to ontology terms, runs differential expression, and builds a Neo4j knowledge graph with Open Targets evidence to rank candidate targets.

    Python

  2. clingenomics clingenomics Public

    Multi-modal clinical genomics interpretation engine: points-based ACMG + PVS1 + RNA evidence + somatic tiering + ClinVar benchmark

    Python

  3. g4-ligand-classifier g4-ligand-classifier Public

    ML classifier for G-quadruplex ligands | ChEMBL data | scaffold-grouped CV AUC 0.94

    Jupyter Notebook

  4. lung-adenocarcinoma-deg-analysis lung-adenocarcinoma-deg-analysis Public

    Differential gene expression analysis and survival testing in lung adenocarcinoma using GEO and TCGA data

    Python

  5. ngs-variant-calling-pipeline ngs-variant-calling-pipeline Public

    Benchmarked NGS variant-calling pipeline (GATK/DeepVariant) with ACMG/AMP 2015 clinical variant classification and automated reporting

    Python

  6. rnaseq-analysis-suite rnaseq-analysis-suite Public

    Bulk RNA-seq, single-cell, and long-read isoform pipelines with ground truth validation

    R