biomedical-data

MCP server for Open Targets data

BERT finetuned on NER downstream tasks

Synthetic biomedical data generator for reproducible benchmarking of feature selection methods in high-dimensional machine learning.

Набор инструментов для обработки радиобиологических Excel‑данных: визуализация опухолевого роста и кожных реакций, статистика, интерактивный GUI (PyQt6). Поддерживается оценка параметров LQ‑модели (α/β) и сравнение экспериментов.

Use ChatGPT, Claude, or Codex to search public dbGaP, UK Biobank, and All of Us metadata through an open-source MCP.

Three different basic data analysis processes of biomedical data for Python. Level: beginner (~200 lines of pure code).

Step1-Step6 preprocessing workflow and final FAERS compound-PT-SOC core graph releases with standardized compounds, MedDRA PT/SOC mapping, and three pruned graph versions.

Multiclass classification of breast cancer subtypes using gene expression profiles. Evaluated and compared multiple models (Logistic Regression, Random Forest, HistGradientBoosting) using classification metrics, confusion matrices, and ROC-AUC analysis with Youden’s J statistic on synthetically generated data

Explainable machine-learning project for malignant vs benign breast cancer classification using the Wisconsin Diagnostic dataset, cross-validation, and SHAP.

This repository contains the data and analysis code for the study "Machine Learning-driven biomarker discovery for stratifying treatment response in tick-borne illness". It investigates the identification of robust and reproducible baseline predictors of treatment response using a stability-aware, multi-method machine learning framework.

Automated harmonization of public omics metadata from GEO, SRA, BioSample, and tabular exports.

Official repository for the paper “Boosting OOD Detection in Biomedical Data with Siamese Neural Networks".

Evidence-backed discovery, qualification and acquisition intelligence for biomedical datasets.

Multiclass classification of breast cancer subtypes using synthetic gene expression data. Refactored code to use a single function for model evaluation across Logistic Regression, Random Forest, and HistGradientBoosting, including metrics and ROC-AUC with Youden’s J statistic.

MSc Bioinformatics project: a Python and Nextflow pipeline for profiling, mapping, validating and harmonising multimodal biomedical data for downstream knowledge-graph integration.

Biomedical graph database application for exploring disease, drug, gene, and anatomy relationships using Python and Kùzu.

MCP server for MyDisease.info API

Reproducible Python QC pipeline for mouse M-mode echocardiography measurements from VisualSonics Vevo systems.

Reproducible data science requirements project for dataset selection, exploratory characterization and analytical-system requirements using the Breast Cancer Wisconsin Diagnostic dataset.