Find open-source science resources

Databank of optimised macromolecular structures. PDB-REDO entries are refined, rebuilt and validated with one consistent protocol using the equivalent entry in the Protein Data Bank and its experimental data. PDB-REDO entries typically have higher structural quality and a better fit to the experimental data.

The European Genome-phenome Archive (EGA) provides long-term storage and controlled sharing of personally identifiable genetic data. The Rega package offers a streamlined and extensible R interface to the EGA API, facilitating the programmatic upload of metadata. GEO-like Excel submission template is provided as a default method of organizing submission metadata.

Highly multiplexed imaging acquires the single-cell expression of selected proteins in a spatially-resolved fashion. These measurements can be visualised across multiple length-scales. First, pixel-level intensities represent the spatial distributions of feature expression with highest resolution. Second, after segmentation, expression values or cell-level metadata (e.g. cell-type information) can be visualised on segmented cell areas. This package contains functions for the visualisation of multiplexed read-outs and cell-level information obtained by multiplexed imaging technologies. The main functions of this package allow 1. the visualisation of pixel-level information across multiple channels, 2. the display of cell-level information (expression and/or metadata) on segmentation masks and 3. gating and visualisation of single cells.

xgt is a command-line tool for programmatic access to the GTDB REST API. It provides four subcommands: search (genome queries with pagination), genome (cards, metadata, taxonomic history), taxon (lineage and genome set retrieval), and diff (per-rank taxonomic comparison between any two GTDB releases). All subcommands support batch input, JSON/CSV/TSV output, file splitting, and automatic retry. Implemented in Rust as a self-contained binary with no runtime dependencies.

Bias factorized, base-resolution deep learning models of chromatin accessibility (chromBPNet).

Lifecycle-Aware Memory (LAM) primitive and benchmark for long-horizon research agents, achieving 65.95% mean reproduction on PaperBench and 94.66% on SurveyBench through Capital Chunk Memory (CCM) with versioned content, structural multi-hop relevance, and provenance-grounded composition; 10 peer-reviewed acceptances at FSE/ICML/TOSEM/AEI/ICoGB (1.3K+ stars)

Programmatic data labeling and weak supervision

Multi-agent system with Parser-Planner-Painter architecture converting `paper.pdf` to editable `poster.pptx`, outperforms GPT-4o with 87% fewer tokens

An interface to the community supported database for amino acid/protein modifications using mass spectrometry.

The PSMatch package helps proteomics practitioners to load, handle and manage Peptide Spectrum Matches. It provides functions to model peptide-protein relations as adjacency matrices and connected components, visualise these as graphs and make informed decision about shared peptide filtering. The package also provides functions to calculate and visualise MS2 fragment ions.

Self-evolving AI research colleague built on OpenClaw with 285+ runtime-adaptive skills across 28+ disciplines, persistent cross-session research memory, and zero-hallucination citation protocols; agent autonomously writes new SKILL.md files based on research patterns without redeployment (828+ stars, MIT License, 2026)

For a convenient overview and download list, visit our model page for this model.

# Instruction For more information, visit our GitHub repository: https://github.com/medfound/medfound

This repository contains GGUF files for gemma4-12b-bioinfo, a fine-tuned Gemma 4 12B model for bioinformatics and computational biology.

gemma4-12b-bioinfo is a fine-tuned Gemma 4 12B instruction model for bioinformatics, genomics, and computational biology question answering.

Provides functions to detect and correct for batch effects in DNA methylation data. The core function is based on latent factor models and can also be used to predict missing values in any other matrix containing real numbers.

modelid = "DuanYi/R3LMHepG2"

Concatenation software for the fast assembly of multi-gene datasets with character set and codon information.

Estimates PCR primer melting temperatures and polymerase-specific annealing temperatures from sequence and buffer inputs, with per-pair QC for hairpins, dimers, and Tm balance. A browser calculator supports interactive single-pair and batch entry (up to 200 pairs) with method comparison and export; a Python library and command-line tool submit the same parameters to the Pepkio Tools API for scripted and pipeline use. Calculator arithmetic for the API client is hosted remotely; sequences are transmitted for programmatic runs while the web interface performs calculations in the browser.

Computes weighed laboratory buffer recipes from target pH, concentration, and volume, accounting for separate preparation and working temperatures when pKa shifts with temperature. Supports calculator mode from dry reagents and stock dilution mode, returning acid and base masses, ionic strength estimates, optional NaCl adjustment, gravimetric and titration routes, and stepwise protocols. A browser calculator supports interactive recipe entry with shareable links; a Python library and command-line tool submit the same parameters to the Pepkio Tools API for scripted and pipeline use. Calculator arithmetic is hosted remotely; the client transmits parameters and returns structured recipe tables, compatibility warnings, and shareable run identifiers.

A merged (ready-to-use) version of microsoft/phi-4 fine-tuned for biological R&D reasoning via QLoRA and evaluated on the Bioalignment Benchmark.

Plans PCR and qPCR master-mix reagent volumes from stock and final concentrations, reaction counts, and pipetting overage, with consolidated totals when several assays are prepared together. A browser calculator supports interactive recipe entry with printable bench sheets; a Python library and command-line tool submit the same parameters to the Pepkio Tools API for scripted and pipeline use. Calculator arithmetic is hosted remotely; the client transmits parameters and returns structured volume tables, dilution warnings, and shareable run identifiers.

Computes laboratory solution preparation parameters—powder mass to weigh, stock and diluent volumes for single dilutions, and multi-step serial concentration tables—with correction for hydrated salts and supplier purity. A browser calculator supports interactive prep planning with saved recipes and shareable links; a Python client and command-line tool submit the same parameters to the Pepkio Tools API for scripted and pipeline use. Calculator arithmetic is hosted remotely; the client transmits parameters and returns structured protocol steps and shareable run identifiers.

The package provides functions to create and use transcript centric annotation databases/packages. The annotation for the databases are directly fetched from Ensembl using their Perl API. The functionality and data is similar to that of the TxDb packages from the GenomicFeatures package, but, in addition to retrieve all gene/transcript models and annotations from the database, ensembldb provides a filter framework allowing to retrieve annotations for specific entries like genes encoded on a chromosome region or transcript models of lincRNA genes. EnsDb databases built with ensembldb contain also protein annotations and mappings between proteins and their encoding transcripts. Finally, ensembldb provides functions to map between genomic, transcript and protein coordinates.

Family of causal genomic foundation models trained on 1T tokens (~6T DNA base pairs) from the Carbon Pretraining Corpus, combining eukaryote genes, mRNA transcripts, and prokaryote genomes with a hybrid text/6-mer tokenizer; Carbon-3B matches or beats Evo2-7B on zero-shot DNA evaluations including sequence recovery, variant effect prediction, and perturbations (Apache 2.0, 201+ stars)

Provides `dplyr` verbs (`mutate`, `select`, `filter`, etc...) supporting `S4Vectors::DataFrame` objects. Importantly, this is achieved without conversion to an intermediate `tibble`. Adds grouping infrastructure to `DataFrame` which is respected by the transformation verbs.

This package provides a periodic table of the elements with support for mass, density and xray/neutron scattering information.

*GenerRNA is a generative pre-trained language model for de novo RNA sequence design. It is a Transformer (decoder-only, GPT-style) model that learns the "language" of RNA from millions of natural sequences and can generate novel, realistic RNA sequences without any structural input, functional…

Plans geometric serial dilution series for molecular biology and biochemistry workflows, rounding transfer volumes to declared pipette ranges and optional 96- or 384-well plate layouts. A browser calculator supports interactive protocol design; a Python client and command-line tool submit the same parameters to the Pepkio Tools API for scripted and pipeline use. Calculator arithmetic is hosted remotely; the client transmits parameters and returns structured step tables and shareable run identifiers.

PathMED is a collection of tools to facilitate precision medicine studies with omics data (e.g. transcriptomics). Among its funcionalities, genesets scores for individual samples may be calculated with several methods. These scores may be used to train machine learning models and to predict clinical features on new data. For this, several machine learning methods are evaluated in order to select the best method based on internal validation and to tune the hyperparameters. Performance metrics and a ready-to-use model to predict the outcomes for new patients are returned.

State-of-the-art RNA 3D folding model developed with Stanford Das Lab and Kaggle competition winners, featuring a 488M-parameter AF3-like architecture with MSA and template-based modeling, enabling structure-driven drug discovery and RNA therapeutics design (NVIDIA-Digital-Bio, Apache 2.0)

esm3-sm-open-v1 is trained on 2.78 billion natural proteins. With synthetic data augmentation, this led to 3.15 billion protein sequences, 236 million protein structures, and 539 million proteins with function annotations, totaling 771 billion tokens.

Original code at (https://github.com/Edoar-do/HuBERT-ECG)

Original code at https://github.com/Edoar-do/HuBERT-ECG

Original code at https://github.com/Edoar-do/HuBERT-ECG

Original code at https://github.com/Edoar-do/HuBERT-ECG

This is a collection of utility functions that allow to perform exploration of and calculations to RNA sequencing data, in a modular, pipe-friendly and tidy fashion.

LLM-native molecular language that represents molecules as explicit graph-based code, enabling LLMs to operate and reason on chemistry directly with 5× lower token cost and ~76-80% accuracy on novel molecules vs ~20% for SMILES; supports small molecules, polymers, and Markush structures with lossless RDKit interconversion and Claude Code/Codex agent skills (AtomFlow, arXiv:2605.16480, 281+ stars, MIT License, 2026)

Foundation model for universal cell segmentation achieving state-of-the-art performance across bacteria, tissue, yeast, cell culture, and diverse imaging modalities (brightfield, fluorescence, phase), with pip-installable inference and Napari plugin (vanvalenlab/Caltech, bioRxiv 2024)

'ggtreeExtra' extends the method for mapping and visualizing associated data on phylogenetic tree using 'ggtree'. These associated data can be presented on the external panels to circular layout, fan layout, or other rectangular layout tree built by 'ggtree' with the grammar of 'ggplot2'.

Apache 2.0 single-cell foundation model family scaling to 3B parameters, pretrained on 266M cell profiles including perturbation data and released with training, embedding, and downstream benchmarking workflows for disease-relevant single-cell tasks (2025)

CliniGuard Vitals NER is a transformer-based clinical Named Entity Recognition model developed by Genzeon Platforms for automated extraction of vital signs, body measurements, and physiological parameters from clinical text.

CliniGuard NER is a clinical Named Entity Recognition model developed by Genzeon Platforms for automated detection and de-identification of Protected Health Information (PHI) and Personally Identifiable Information (PII) in clinical text.

This model card provides an overview of the intended use of the ESMC SAE models and examples of how to access them, but it does not have a specific model or model weights. To access each SAE model collection, use the links below:

This model card provides an overview of the intended use of the ESMC SAE models and examples of how to access them, but it does not have a specific model or model weights. To access each SAE model collection, use the links below:

This set of model weights was released with the GitHub-compatible esm package format. The models here are kept for backwards compatibility, but we recommend you use the HuggingFace-compatible model weights at biohub/ESMC-6B (or biohub/ESMC-300M / biohub/ESMC-600M) instead.

ESMC is a state-of-the-art protein language model that has learned the rules of protein biology from training on billions of protein sequences. ESMC provides representations of proteins enabling novel AI applications from therapeutic protein engineering to unlocking basic insights into protein…

ESMC is a state-of-the-art protein language model that has learned the rules of protein biology from training on billions of protein sequences. ESMC provides representations of proteins enabling novel AI applications from therapeutic protein engineering to unlocking basic insights into protein…

ESMC is a state-of-the-art protein language model that has learned the rules of protein biology from training on billions of protein sequences. ESMC provides representations of proteins enabling novel AI applications from therapeutic protein engineering to unlocking basic insights into protein…