# Toolome

> Curated, hand-verified directory of AI tools for biology (188 tools). Each tool page lists URL, GitHub, paper, license and pricing.

## Single-cell & foundation models

- [scGPT](https://toolome.bio/tools/scgpt/): Generative pretrained transformer foundation model for single-cell multi-omics.
- [Geneformer](https://toolome.bio/tools/geneformer/): Transformer pretrained on ~30M single-cell transcriptomes for network biology predictions.
- [scvi-tools](https://toolome.bio/tools/scvi-tools/): Probabilistic deep generative models for single-cell omics (scVI, totalVI, scANVI...).
- [CellTypist](https://toolome.bio/tools/celltypist/): Automated cell type annotation with logistic regression models trained on curated atlases.
- [Scanpy](https://toolome.bio/tools/scanpy/): Scalable Python toolkit for single-cell gene expression analysis.
- [Seurat](https://toolome.bio/tools/seurat/): R toolkit for single-cell genomics, integration and multimodal analysis.
- [scFoundation](https://toolome.bio/tools/scfoundation/): 100M-parameter foundation model trained on 50M human single-cell transcriptomes.
- [UCE (Universal Cell Embeddings)](https://toolome.bio/tools/uce-universal-cell-embeddings/): Cross-species zero-shot universal cell embedding foundation model.
- [CellRank](https://toolome.bio/tools/cellrank/): Cell fate mapping using Markov chains on single-cell data.
- [scVelo](https://toolome.bio/tools/scvelo/): RNA velocity estimation with dynamical modeling.
- [MuData / muon](https://toolome.bio/tools/mudata--muon/): Multimodal omics framework for single-cell multi-omics in Python.
- [ArchR](https://toolome.bio/tools/archr/): Scalable single-cell chromatin accessibility (scATAC-seq) analysis in R.
- [CellOracle](https://toolome.bio/tools/celloracle/): In silico gene perturbation via GRN inference from single-cell multi-omics.
- [GEARS](https://toolome.bio/tools/gears/): Graph-based prediction of transcriptional outcomes of genetic perturbations.
- [Harmony](https://toolome.bio/tools/harmony/): Fast, sensitive batch integration of single-cell data.
- [PIE (Arc Institute)](https://toolome.bio/tools/pie/): Virtual-cell model using curated biological knowledge to predict perturbation responses in unseen contexts.
- [scBasset](https://toolome.bio/tools/scbasset/): Sequence-based deep learning model for single-cell ATAC-seq representation and TF activity.
- [State](https://toolome.bio/tools/state/): Arc Institute virtual-cell model predicting transcriptomic responses to perturbations across cell contexts.
- [TranscriptFormer](https://toolome.bio/tools/transcriptformer/): CZI cross-species generative single-cell foundation model trained on 112M cells from 12 species.

## Spatial omics

- [Tangram](https://toolome.bio/tools/tangram/): Maps single-cell profiles onto spatial data with deep learning.
- [PASTE](https://toolome.bio/tools/paste/): Optimal-transport alignment and integration of spatial transcriptomics slices.
- [moscot](https://toolome.bio/tools/moscot/): Multi-omics single-cell optimal transport for space and time mapping.
- [cell2location](https://toolome.bio/tools/cell2location/): Bayesian mapping of fine-grained cell types in spatial transcriptomics.
- [Squidpy](https://toolome.bio/tools/squidpy/): Spatial single-cell analysis toolkit in the scverse ecosystem.
- [SpatialData](https://toolome.bio/tools/spatialdata/): Unified data framework for spatial omics technologies.
- [STAGATE](https://toolome.bio/tools/stagate/): Graph attention autoencoder for spatial domain identification.
- [GraphST](https://toolome.bio/tools/graphst/): Graph self-supervised contrastive learning for spatial transcriptomics.
- [BayesSpace](https://toolome.bio/tools/bayesspace/): Bayesian clustering and resolution enhancement of spatial transcriptomics.
- [RCTD (spacexr)](https://toolome.bio/tools/rctd-spacexr/): Robust cell type decomposition for spatial transcriptomics.
- [Giotto Suite](https://toolome.bio/tools/giotto-suite/): Comprehensive R toolbox for spatial data analysis and visualization.
- [Nicheformer](https://toolome.bio/tools/nicheformer/): Foundation model pretrained on dissociated and spatial single-cell data.
- [Alpha Spatial](https://toolome.bio/tools/alpha-spatial/): Language-conditioned model that predicts cell-level spatial gene expression from histology across tissues and gene panels.

## Protein structure & design

- [AlphaFold Server (AlphaFold 3)](https://toolome.bio/tools/alphafold-server-alphafold-3/): Structure prediction of proteins, nucleic acids, ligands and their complexes; free web server and code, weights on request for non-commercial use.
- [AlphaFold DB](https://toolome.bio/tools/alphafold-db/): 200M+ predicted protein structures from DeepMind and EMBL-EBI.
- [RFdiffusion](https://toolome.bio/tools/rfdiffusion/): Diffusion model for de novo protein backbone design and binder design.
- [ProteinMPNN](https://toolome.bio/tools/proteinmpnn/): Deep-learning inverse folding: designs sequences for given backbones.
- [ESM3 / EvolutionaryScale](https://toolome.bio/tools/esm3--evolutionaryscale/): Multimodal generative protein language model over sequence, structure and function (ESM3; also ESM C embeddings).
- [ESM-2 / ESMFold](https://toolome.bio/tools/esm-2--esmfold/): Protein language models and single-sequence structure prediction (Meta repo archived; models still widely used via Hugging Face).
- [Boltz](https://toolome.bio/tools/boltz/): Open biomolecular interaction models (Boltz-1, Boltz-2) for complex structure and binding-affinity prediction.
- [Chai-1](https://toolome.bio/tools/chai-1/): Multimodal foundation model for molecular structure prediction.
- [OpenFold](https://toolome.bio/tools/openfold/): Trainable open-source reproduction of AlphaFold2 in PyTorch.
- [ColabFold](https://toolome.bio/tools/colabfold/): Fast, accessible protein folding with MMseqs2 + AlphaFold in Colab.
- [RoseTTAFold All-Atom](https://toolome.bio/tools/rosettafold-all-atom/): Generalized biomolecular modeling for proteins, nucleic acids, small molecules.
- [BindCraft](https://toolome.bio/tools/bindcraft/): One-shot de novo binder design pipeline with AF2 hallucination.
- [Foldseek](https://toolome.bio/tools/foldseek/): Fast and accurate protein structure search.
- [Protenix](https://toolome.bio/tools/protenix/): Trainable PyTorch reproduction of AlphaFold 3 by ByteDance.
- [Tamarind Bio](https://toolome.bio/tools/tamarind-bio/): No-code web platform running 100+ protein and molecular ML models.
- [AlphaProtein Novo](https://toolome.bio/tools/alphaprotein-novo/): Google DeepMind diffusion pipeline for de novo enzyme design via catalytic motif scaffolding.
- [Prompteina](https://toolome.bio/tools/prompteina/): Text-to-structure all-atom protein design: generate monomers and binders from natural-language prompts.
- [IgFold](https://toolome.bio/tools/igfold/): Fast antibody structure prediction using AntiBERTy embeddings and graph networks.
- [AbLang](https://toolome.bio/tools/ablang/): Antibody language model for heavy/light chains, used for residue restoration and embeddings.
- [AbLang-2](https://toolome.bio/tools/ablang2/): Paired-chain antibody language model trained to reduce germline bias.
- [ImmuneBuilder](https://toolome.bio/tools/immunebuilder/): Deep learning structure prediction for antibodies, nanobodies and TCRs (ABodyBuilder2 etc.).
- [AntiBERTy](https://toolome.bio/tools/antiberty/): BERT-style antibody language model trained on 558M OAS sequences.
- [RFantibody](https://toolome.bio/tools/rfantibody/): Structure-based de novo antibody and nanobody design with fine-tuned RFdiffusion, ProteinMPNN and RF2.
- [BioPhi](https://toolome.bio/tools/biophi/): Antibody humanization (Sapiens) and humanness evaluation (OASis) platform.
- [AlphaFold 2](https://toolome.bio/tools/alphafold2/): DeepMind's open-source protein structure prediction model (monomer and multimer).
- [BioEmu](https://toolome.bio/tools/bioemu/): Generative model emulating protein equilibrium structural ensembles.
- [Proteina](https://toolome.bio/tools/proteina/): NVIDIA flow-matching model for large-scale protein backbone generation.
- [La-Proteina](https://toolome.bio/tools/la-proteina/): NVIDIA latent flow model for joint atomistic protein sequence-structure generation.
- [ProGen2](https://toolome.bio/tools/progen2/): Protein language models (up to 6.4B) for sequence generation and fitness prediction.
- [RFdiffusion2](https://toolome.bio/tools/rfdiffusion2/): Atom-level enzyme active-site scaffolding with diffusion, no sequence-position constraints.
- [BoltzGen](https://toolome.bio/tools/boltzgen/): Open generative model for designing protein and peptide binders against diverse targets.
- [CHORUS](https://toolome.bio/tools/chorus/): Sequence-only pLM ensemble predicting membrane-protein interaction interfaces; proteome-scale atlas covering 80k+ reviewed TM proteins.

## Genomics & DNA models

- [Evo 2](https://toolome.bio/tools/evo-2/): Genome language model across all domains of life with 1M-token context.
- [AlphaGenome](https://toolome.bio/tools/alphagenome/): DeepMind model predicting regulatory effects of DNA variants at base resolution.
- [Enformer](https://toolome.bio/tools/enformer/): Predicts gene expression from sequence with long-range interactions.
- [SpliceAI](https://toolome.bio/tools/spliceai/): Deep learning prediction of splicing effects from pre-mRNA sequence (repo archived).
- [Nucleotide Transformer](https://toolome.bio/tools/nucleotide-transformer/): DNA foundation models pretrained on 3,200+ human and 850 species genomes.
- [HyenaDNA](https://toolome.bio/tools/hyenadna/): Long-range genomic foundation model at single-nucleotide resolution.
- [DNABERT-2](https://toolome.bio/tools/dnabert-2/): Efficient multi-species genome foundation model.
- [Borzoi](https://toolome.bio/tools/borzoi/): Predicts RNA-seq coverage from DNA sequence.
- [AlphaMissense](https://toolome.bio/tools/alphamissense/): Proteome-wide pathogenicity predictions for 71M missense variants (repo archived; precomputed scores available).
- [Caduceus](https://toolome.bio/tools/caduceus/): Bidirectional, reverse-complement-equivariant DNA language model.
- [SpliceAI2](https://toolome.bio/tools/spliceai2/): Illumina's successor to SpliceAI: predicts splice-site usage and full transcript isoforms from sequence.
- [Carbon-A](https://toolome.bio/tools/carbon-a/): 1.2B DNA model that annotates protein-coding genes across eukaryotes; ships a 48K-assembly annotation database.
- [Basenji](https://toolome.bio/tools/basenji/): Sequence-to-function CNNs predicting chromatin accessibility, histone marks and expression tracks from DNA.
- [ChromBPNet](https://toolome.bio/tools/chrombpnet/): Bias-corrected base-resolution deep learning models of ATAC/DNase-seq chromatin accessibility and TF motifs.
- [BPNet](https://toolome.bio/tools/bpnet/): Base-resolution neural network for TF binding profiles (ChIP-nexus/ChIP-seq) and motif syntax discovery.
- [Sei](https://toolome.bio/tools/sei/): Predicts 21,907 chromatin profiles from sequence and maps variants to regulatory sequence classes.
- [Selene](https://toolome.bio/tools/selene/): PyTorch library for training and applying sequence-level deep learning models (DeepSEA-style).
- [EVE](https://toolome.bio/tools/eve/): Unsupervised evolutionary VAE for predicting pathogenicity of missense variants.
- [PrimateAI-3D](https://toolome.bio/tools/primateai-3d/): Missense variant pathogenicity model trained on primate population variation and 3D protein structure.
- [GPN-MSA](https://toolome.bio/tools/gpn-msa/): Genomic language model over multi-species alignments for genome-wide variant effect prediction.
- [Evo](https://toolome.bio/tools/evo/): 7B genomic foundation model (StripedHyena) for prokaryotic DNA, RNA and protein design at single-nucleotide resolution.

## Drug discovery & molecules

- [RDKit](https://toolome.bio/tools/rdkit/): Open-source cheminformatics and machine learning toolkit.
- [DeepChem](https://toolome.bio/tools/deepchem/): Deep learning toolkit for drug discovery, materials and biology.
- [DiffDock](https://toolome.bio/tools/diffdock/): Diffusion generative model for molecular docking.
- [REINVENT 4](https://toolome.bio/tools/reinvent-4/): AstraZeneca's generative molecular design framework.
- [Chemprop](https://toolome.bio/tools/chemprop/): Message passing neural networks for molecular property prediction.
- [TorchDrug](https://toolome.bio/tools/torchdrug/): PyTorch platform for drug discovery ML.
- [AutoDock Vina](https://toolome.bio/tools/autodock-vina/): Fast, widely used open-source molecular docking engine.
- [OpenMM](https://toolome.bio/tools/openmm/): High-performance molecular dynamics toolkit with GPU support.
- [NVIDIA BioNeMo](https://toolome.bio/tools/nvidia-bionemo/): NVIDIA platform and NIMs for training and deploying biology foundation models.
- [Rowan](https://toolome.bio/tools/rowan/): Cloud computational chemistry platform with ML-accelerated simulations.

## Bio LLM agents & copilots

- [Biomni](https://toolome.bio/tools/biomni/): General-purpose biomedical AI agent with 150+ tools and databases.
- [Edison Scientific Platform (FutureHouse)](https://toolome.bio/tools/futurehouse-platform/): AI scientist agents (formerly FutureHouse Crow, Falcon, Owl, Phoenix) for literature search and chemistry.
- [PaperQA2](https://toolome.bio/tools/paperqa2/): High-accuracy RAG agent over scientific papers.
- [BioChatter](https://toolome.bio/tools/biochatter/): Framework for LLM-based conversational biomedical research.
- [Elicit](https://toolome.bio/tools/elicit/): AI research assistant for literature search and data extraction.
- [Consensus](https://toolome.bio/tools/consensus/): AI search engine over peer-reviewed research.
- [Benchling](https://toolome.bio/tools/benchling/): R&D cloud with ELN, molecular biology and AI tooling.
- [Latch Bio](https://toolome.bio/tools/latch-bio/): Cloud platform for bioinformatics workflows and data.
- [Coscientist](https://toolome.bio/tools/coscientist/): GPT-4-driven agent that plans and executes chemistry experiments on lab hardware.
- [ChemCrow](https://toolome.bio/tools/chemcrow/): LLM chemistry agent augmented with 18 expert tools for synthesis and molecule tasks.
- [PyLabRobot](https://toolome.bio/tools/pylabrobot/): Hardware-agnostic Python interface for liquid-handling robots and lab automation.
- [BioDiscoveryAgent](https://toolome.bio/tools/biodiscoveryagent/): LLM agent that designs genetic perturbation experiments (e.g. CRISPR screens).
- [Robin](https://toolome.bio/tools/robin/): FutureHouse multi-agent system automating hypothesis generation and data analysis for drug discovery.
- [Aviary](https://toolome.bio/tools/aviary/): Gymnasium for training language agents on scientific tasks (molecular cloning, literature QA).
- [Enrich.AI](https://toolome.bio/tools/enrich-ai/): Conversational ReAct agent for gene-set enrichment: 220+ libraries, Europe PMC literature grounding, interactive plots and human-in-the-loop refinement.

## Cloud labs & wet-lab services

- [Adaptyv Bio](https://toolome.bio/tools/adaptyv-bio/): Cloud lab for protein expression and binding characterization, binder design competitions.
- [Emerald Cloud Lab](https://toolome.bio/tools/emerald-cloud-lab/): Remote-controlled cloud laboratory running 200+ instrument types.
- [Ginkgo Bioworks](https://toolome.bio/tools/ginkgo-bioworks/): Cell programming platform and automated lab services.
- [Twist Bioscience](https://toolome.bio/tools/twist-bioscience/): Synthetic DNA, gene fragments and antibody discovery services.
- [Strateos](https://toolome.bio/tools/strateos/): Robotic cloud lab automation for life sciences.
- [Science Exchange](https://toolome.bio/tools/science-exchange/): Marketplace for outsourced R&D and CRO services.
- [Opentrons](https://toolome.bio/tools/opentrons/): Open-source liquid-handling robots with a Python Protocol API and AI protocol generation.

## Imaging & pathology

- [Cellpose](https://toolome.bio/tools/cellpose/): Generalist deep learning model for cellular segmentation.
- [StarDist](https://toolome.bio/tools/stardist/): Star-convex object detection for nuclei segmentation.
- [napari](https://toolome.bio/tools/napari/): Fast, interactive multi-dimensional image viewer for Python.
- [QuPath](https://toolome.bio/tools/qupath/): Open-source software for digital pathology and whole slide image analysis.
- [UNI](https://toolome.bio/tools/uni/): General-purpose self-supervised foundation model for computational pathology.
- [CONCH](https://toolome.bio/tools/conch/): Vision-language foundation model for histopathology.
- [Prov-GigaPath](https://toolome.bio/tools/prov-gigapath/): Whole-slide digital pathology foundation model.
- [CellProfiler](https://toolome.bio/tools/cellprofiler/): Image analysis software for high-throughput cell phenotyping.
- [micro-sam](https://toolome.bio/tools/micro-sam/): Segment Anything for microscopy.
- [BiomedParse](https://toolome.bio/tools/biomedparse/): Foundation model for joint segmentation, detection and recognition across nine biomedical imaging modalities.

## Datasets & benchmarks

- [CZ CELLxGENE](https://toolome.bio/tools/cz-cellxgene/): Curated single-cell data portal and Census API with 100M+ cells.
- [Human Cell Atlas Data Portal](https://toolome.bio/tools/human-cell-atlas-data-portal/): Data portal of the Human Cell Atlas consortium.
- [Tahoe-100M](https://toolome.bio/tools/tahoe-100m/): 100M-cell single-cell drug perturbation atlas.
- [Protein Data Bank (RCSB)](https://toolome.bio/tools/protein-data-bank-rcsb/): Archive of experimentally determined 3D biomolecular structures.
- [UniProt](https://toolome.bio/tools/uniprot/): Comprehensive resource of protein sequence and function.
- [ProteinGym](https://toolome.bio/tools/proteingym/): Benchmarks for protein fitness and mutation effect prediction.
- [Therapeutics Data Commons](https://toolome.bio/tools/therapeutics-data-commons/): ML-ready datasets and benchmarks for therapeutics.
- [CASP](https://toolome.bio/tools/casp/): Critical Assessment of protein Structure Prediction community experiment.
- [Open Problems in Single-Cell](https://toolome.bio/tools/open-problems-in-single-cell/): Living benchmarks for single-cell analysis tasks.
- [PoseBusters](https://toolome.bio/tools/posebusters/): Plausibility checks and benchmark for docking/structure poses.
- [Polaris](https://toolome.bio/tools/polaris/): Benchmarking platform for ML in drug discovery.
- [10x Genomics Datasets](https://toolome.bio/tools/10x-genomics-datasets/): Public single-cell and spatial (Visium, Xenium) example datasets.
- [Hugging Face Bio models](https://toolome.bio/tools/hugging-face-bio-models/): Hub listing of biology-tagged models.

## Sequencing & variant calling

- [DeepVariant](https://toolome.bio/tools/deepvariant/): Google deep learning variant caller for Illumina, PacBio HiFi and ONT data.
- [Dorado](https://toolome.bio/tools/dorado/): Oxford Nanopore's production basecaller: transformer/LSTM models turning raw nanopore signal into reads, with modified-base calling and duplex.
- [Bonito](https://toolome.bio/tools/bonito/): Oxford Nanopore's research basecaller for training and testing new basecalling models.
- [Medaka](https://toolome.bio/tools/medaka/): Neural-network consensus polishing and variant calling for nanopore assemblies.
- [Remora](https://toolome.bio/tools/remora/): Models for detecting modified bases (e.g. 5mC, 6mA) from nanopore signal.
- [Clair3](https://toolome.bio/tools/clair3/): Deep-learning germline small-variant caller for long reads (ONT, PacBio) and short reads.
- [DeepConsensus](https://toolome.bio/tools/deepconsensus/): Transformer that corrects PacBio HiFi consensus reads, increasing yield and accuracy.
- [PEPPER-Margin-DeepVariant](https://toolome.bio/tools/pepper-margin-deepvariant/): Haplotype-aware variant calling pipeline for nanopore reads combining PEPPER, Margin and DeepVariant.
- [Sniffles2](https://toolome.bio/tools/sniffles2/): Fast structural-variant caller for long reads with population-scale and mosaic SV detection.

## Proteomics & metabolomics

- [DIA-NN](https://toolome.bio/tools/dia-nn/): Neural-network-powered data-independent acquisition (DIA) proteomics processing software.
- [AlphaPeptDeep](https://toolome.bio/tools/alphapeptdeep/): Deep learning framework predicting peptide MS2 spectra, retention time and CCS for proteomics.
- [AlphaDIA](https://toolome.bio/tools/alphadia/): Feature-free DIA proteomics search engine with transfer learning from AlphaPeptDeep.
- [MS2PIP](https://toolome.bio/tools/ms2pip/): Predicts peptide fragmentation spectra (and with DeepLC/IM2Deep) for spectral libraries and rescoring.
- [DeepLC](https://toolome.bio/tools/deeplc/): Deep learning retention-time predictor that generalizes to unseen peptide modifications.
- [Casanovo](https://toolome.bio/tools/casanovo/): Transformer for de novo peptide sequencing directly from MS/MS spectra.
- [InstaNovo](https://toolome.bio/tools/instanovo/): Transformer and diffusion models for de novo peptide sequencing from tandem mass spectra.
- [SIRIUS](https://toolome.bio/tools/sirius/): Metabolomics software for molecular formula and structure annotation from MS/MS with CSI:FingerID and CANOPUS.
- [MS2DeepScore](https://toolome.bio/tools/ms2deepscore/): Siamese neural network predicting structural similarity between MS/MS spectra for metabolomics.
- [matchms](https://toolome.bio/tools/matchms/): Python library for importing, cleaning and comparing mass spectra for metabolomics.
- [DreaMS](https://toolome.bio/tools/dreams/): Self-supervised transformer foundation model for tandem mass spectra of small molecules.
- [MZmine](https://toolome.bio/tools/mzmine/): Open-source platform for LC-MS and imaging-MS data processing in metabolomics.

## CRISPR & gene editing

- [CRISPOR](https://toolome.bio/tools/crispor/): Web tool for CRISPR guide selection with on-target efficiency and off-target scoring for 500+ genomes.
- [CHOPCHOP](https://toolome.bio/tools/chopchop/): Web tool for designing CRISPR-Cas9, Cas12 and TALEN targets with efficiency predictions.
- [Rule Set 3 (rs3)](https://toolome.bio/tools/rule-set-3/): Broad GPP model for SpCas9 sgRNA on-target activity, used in CRISPick.
- [PRIDICT](https://toolome.bio/tools/pridict/): Attention-based model predicting prime editing efficiency to design pegRNAs.
- [CRISPResso2](https://toolome.bio/tools/crispresso2/): Analysis of genome-editing outcomes from amplicon sequencing, including base and prime editing.
- [OpenCRISPR-1](https://toolome.bio/tools/opencrispr/): AI-designed Cas9-like gene editor from Profluent's language models, released openly.

## RNA structure & design

- [RhoFold+](https://toolome.bio/tools/rhofold/): RNA language-model-based method for 3D RNA structure prediction.
- [RNA-FM](https://toolome.bio/tools/rna-fm/): RNA foundation model pretrained on 23M non-coding RNAs for structure and function tasks.
- [RibonanzaNet](https://toolome.bio/tools/ribonanzanet/): RNA foundation model trained on Ribonanza chemical-mapping data for structure and reactivity.
- [EternaFold](https://toolome.bio/tools/eternafold/): RNA secondary structure model trained via multitask learning on Eterna player data.
- [SPOT-RNA](https://toolome.bio/tools/spot-rna/): Deep learning RNA secondary structure prediction including non-canonical pairs and pseudoknots.
- [UFold](https://toolome.bio/tools/ufold/): Fast image-like U-Net for RNA secondary structure prediction.
- [ViennaRNA](https://toolome.bio/tools/viennarna/): Standard RNA secondary structure prediction and design package (RNAfold, RNAinverse).
- [gRNAde](https://toolome.bio/tools/grnade/): Geometric deep learning for 3D RNA inverse design (sequence design from backbone).

## Microbiome & metagenomics

- [MetaPhlAn 4](https://toolome.bio/tools/metaphlan/): Marker-gene-based taxonomic profiling of metagenomes, including uncharacterized species.
- [HUMAnN 3](https://toolome.bio/tools/humann/): Functional profiling of metagenomes and metatranscriptomes (gene families and pathways).
- [Kraken 2](https://toolome.bio/tools/kraken2/): Fast k-mer based taxonomic classification of metagenomic reads.
- [GTDB-Tk](https://toolome.bio/tools/gtdb-tk/): Assigns GTDB taxonomy to bacterial and archaeal genomes and MAGs.
- [geNomad](https://toolome.bio/tools/genomad/): Neural-network and marker-based identification of viruses and plasmids in sequencing data.
- [CheckM2](https://toolome.bio/tools/checkm2/): Machine-learning estimation of completeness and contamination of metagenome-assembled genomes.
- [SemiBin2](https://toolome.bio/tools/semibin/): Self-supervised deep learning metagenomic binning for short and long reads.
- [VAMB](https://toolome.bio/tools/vamb/): Variational autoencoder for metagenomic binning.
- [QIIME 2](https://toolome.bio/tools/qiime2/): Extensible platform for microbiome (amplicon and shotgun) data science.

## Clinical & biomedical LLMs

- [Meditron](https://toolome.bio/tools/meditron/): Open medical LLMs (7B/70B) adapted from Llama 2 on curated medical corpora.
- [MedGemma](https://toolome.bio/tools/medgemma/): Google's open Gemma 3 variants for medical text and image comprehension.
- [LLaVA-Med](https://toolome.bio/tools/llava-med/): Biomedical vision-language assistant trained on PubMed Central figure-caption data.
- [BioMistral](https://toolome.bio/tools/biomistral/): Open 7B medical LLM based on Mistral, further pretrained on PubMed Central.