Changelog

Added ESM-C for generating protein embeddings and optional masked-token logits from ESM Cambrian models.

Added command palette navigation and onboarding progress so users can jump to tools, docs, jobs, and account actions more quickly.

Added searchable tool settings for faster configuration on tools with many parameters.

Added Proteo-R1 reasoning-guided antibody CDR co-design for antibody-antigen complexes, with ranked PDB structures, sequence tables, CDR prediction JSON, and confidence artifacts.

Added ESMFold2 for protein and protein-complex structure prediction with confidence outputs and pLDDT-colored structure viewing.

Improved PDB2PQR processing reliability for structures that previously failed during preparation.

Improved AutoDock Vina result filenames so downloaded ligand poses stay clearly associated with their source ligands.

Improved metered design jobs to return completed partial outputs when a run ends early, including candidate structures from BindCraft.

Added integrated support ticket submission and email delivery for smoother help requests.

Moved help content into the main documentation experience for a more consistent support workflow.

Added metered runtime billing for long-running design tools so credit use better reflects actual compute time.

Improved ColabDock, DiffDock-L, and SurfDock reliability with longer timeouts, updated pricing, and pose parsing fixes.

Improved job polling and output activation so tool results appear more reliably when runs complete.

Improved the server tool workbench output flow for clearer transitions between empty, running, and completed states.

Improved AutoDock Vina file access performance for repeated result downloads.

Added validation for HyperMPNN residue settings.

Improved RFantibody defaults, PDB validation, insertion-code handling, and output capture for antibody and nanobody binder design.

Improved DLKcat with paired enzyme-substrate inputs and fuller result outputs.

Improved LightDock controls, documentation, and result fidelity.

Added minimum-sequence validation to MUSCLE5.

Added plan-based controls for additional AutoDock Vina poses.

Improved empty states for design tools with direct links into each tool's documentation.

Added downloadable pose complexes and CIF complex files to the structure viewer.

Improved structure conversion so PDB bond records are preserved when generating CIF outputs.

Improved BindCraft runtime pricing and candidate handling.

Improved RFantibody PDB upload detection.

Improved job history filter labels and structure pricing display accuracy.

Strengthened subscription change handling in billing.

Improved the tool workbench layout across server-side tools.

Improved Mol* structure layer labels and refreshed the viewer bundle for better structure inspection.

Improved settings rendering performance and slider responsiveness.

Improved AutoDock Vina support for multi-record MOL2 ligand inputs.

Added Carbon for generating, scoring, and comparing canonical DNA sequences with Carbon language models.

Added PocketXMol for generating docking poses, drug-like molecules, and peptides inside protein binding pockets.

Added metered credit reservations for molecular dynamics tools.

Improved AutoDock Vina receptor atom-type handling.

Added input validation for ThermoMPNN chain selection, RAxML-NG alignments, and DiffDock-L PDB inputs.

Improved tool settings and shared app surfaces for faster tool setup.

Reduced molecular input logging to better protect sensitive structures and sequences.

Added IQ-TREE bootstrap sequence validation.

Improved PROPKA 3 PDB charge validation for molecular dynamics inputs.

Corrected the Extinction Coefficient interface text.

Improved settings animations and notice messages across tool pages.

Expanded BindCraft protocol settings for finer control over binder design runs.

Improved AlphaFlow runtime reliability for DeepSpeed-backed ensemble generation.

Improved OpenFold 3 fallback behavior for prediction runs.

Reduced duplicate job polling on tool pages.

Improved PoseBusters validation outputs and runtime behavior.

Added HADDOCK3 support for heterogen topology files.

Added input guardrails for AbLang RNA submissions and PDBQT-style AutoDock Vina receptor uploads.

Improved credit calculations for Chai-1 and IgDesign.

Improved workflow templates so generated workflows launch with sound tool and data connections.

Added tiered GPU routing for ColabDock.

Improved OpenMM validation for ligand-only structures and ligand parameterization failures.

Added paired FASTA validation for AbLang-2 and antibody-sequence validation for BioPhi.

Added compatibility warnings for AutoDock Vina receptor uploads.

Improved the submit bar credit display and tool settings controls.

Added tiered GPU routing and larger-job support for RFDiffusion 2.

Added a credit calculator for TLimmuno2.

Improved ImmuneBuilder TCR fallback behavior.

Improved RFDiffusion runtime reliability.

Improved concurrent job lifecycle handling.

Improved support request email replies.

Improved billing invoice actions, including invoice visibility for expired subscriptions.

Improved DFMDock execution reliability by resolving a sampling runtime failure.

Added LMI4Boltz for low-memory Boltz inference on biomolecular structure and affinity prediction.

Added PoseBusters for validating generated and docked ligand poses with structure, chemistry, and reference-pose checks.

Improved HADDOCK3 AIR restraint validation and multichain documentation.

Improved Aggrescan3D result rendering.

Improved workflow credit estimates and fan-out behavior for multi-FASTA inputs.

Improved dataset and database browsing.

Added empty-SMILES upload validation.

Improved saved-job input extraction for ANARCI.

Improved Aggrescan3D execution reliability.

Improved paid signup and credit activation flow.

Improved structure viewer defaults for Mol* outputs.

Reduced duplicate job polling.

Improved AlphaFlow ESMFlow behavior and documentation.

Improved HyperMPNN settings and documentation.

Improved ProstT5 translations and embeddings.

Added citation popups to articles.

Improved molecular descriptor outputs for small-molecule tools.

Improved TXT to FASTA conversion behavior.

Improved AutoDock Vina receptor advisory messages.

Added tool-page schema metadata and selected guide redirects for better discovery.

Increased the free daily job limit.

Improved workflow execution hardening and internal route protection.

Improved Boltz-2 MSA URL handling and telemetry protections.

Rendered tool documentation directly for crawlers to improve search visibility.

Added workflow plan limits and workflow template category filters.

Added plan-based controls for AlphaFold2 MSA usage.

Improved workflow alpha stability and execution hardening.

Improved OpenFold 3 prediction reliability.

Improved structure viewer pose loading.

Simplified signup and workflow authentication flows.

Added self-serve account deletion.

Rebuilt authenticated onboarding.

Added clearer annual-plan monthly reference pricing.

Added validation for invalid paired AntiFold inputs.

Improved SmrTNet output behavior.

Improved TXT to FASTA detection for multiline sequences.

Added validation to block single-sequence MAFFT submissions.

Launched Lite self-serve pricing plans.

Added a ProFam credit calculator.

Improved docking tool configurations and related-tool recommendations.

Stabilized job launch persistence and polling.

Improved billing subscription recovery and pricing help content.

Improved job history tool links.

Improved workflow quality-of-life behavior.

Updated public pricing and policy copy.

Added a databases browser and provider registry for searchable external data sources.

Overhauled workflow execution and the workflow editor.

Improved docking and screening tools with more faithful settings, validation, and outputs.

Added repair for malformed PDB CONECT records.

Improved billing access flows and shared app surfaces.

Improved subscription lifecycle handling.

Improved free tool conversions.

Added workflow node category filters and split the node panel into data and tool tabs.

Added FASTA identifier validation to FastTree.

Improved Clustal Omega FASTA normalization.

Improved workflow auto-layout spacing and default viewport behavior.

Improved workflow inputs by reusing the shared external fetcher.

Improved AI assistant batch cancellation and upload-limit handling.

Improved Aggrescan3D, SurfDock, AntiFold, ColabDock, AllMetal3D, and ESMFold runtime behavior and outputs.

Improved AlphaGenome chromosome validation.

Improved Boltz-2 multi-component ligand affinity handling.

Centralized tool-page citations for more consistent academic references.

Improved TXT to FASTA documentation for manual conversion workflows.

Improved TLimmuno2 runtime behavior and outputs.

Improved Admetica result behavior.

Improved RNAplfold opening energy labels.

Improved HMMER and SurfDock behavior.

Added protein-backbone validation to structure-conditioned tools.

Added guardrails for wildcard query atoms in AutoDock Vina ligands.

Improved PDB2PQR handling for hydrogenated inputs.

Added guardrails for SMILES query atoms in docking tools.

Improved gmx_MMPBSA tool image handling.

Updated the Protenix title.

Added precomputed MSA support to Protenix.

Improved BindCraft, DockQ, EvoDiff, mBER, and SPRINT runtime behavior and outputs.

Added validation for malformed PDBQT ligands before AutoDock Vina runs.

Improved mBER outputs and defaults.

Improved Primer3 behavior.

Improved AI assistant artifact loading and response caching.

Updated tool documentation and removed outdated related-tool recommendations.

Improved multiple tool pages with clearer settings and results.

Added a chain-count guardrail to DiffDock-L for multimeric receptors.

Improved PDB validation for CHARMM-GUI structures with misaligned residue columns.

Consolidated glossary references into the main docs.

Added DFMDock for diffusion-based protein-protein docking with confidence-ranked complex predictions.

Added batch workflows that can run multiple tools and list-mode inputs.

Improved Boltz-2 validation for large ligands.

Improved RFDiffusion3, HADDOCK3, AutoDock Vina, and PDB Fixer tool behavior.

Updated pricing, legal, help center, guide, navigation, and app layout surfaces.

Added case-study pages for customer research stories.

Improved DiffDock-L receptor parsing for tokenized PDB files.

Added ProGen2 for de novo protein sequence generation from plain-text prompts.

Added pySCA for statistical coupling analysis of co-evolving residue sectors.

Improved Boltz-2, AutoDock-GPU, AutoDock Vina, and SurfDock input validation and execution.

Improved IgGM pricing, timeout handling, and conflicting epitope-setting validation.

Improved tool submission flow, job output components, and workflow logic.

Added shared file previews for fetched and uploaded files.

Improved molecular dynamics credit calculators for GROMACS and OpenMM.

Added tiered GPU dispatch and atom-based credit calculation for Orb-v3.

Improved PepMLM citation links and batch output sequence handling.

Restored ViennaRNA tools into the analysis catalog and improved RNA fold visualization, CLI parsing, docs, and result fidelity across the RNA suite.

Expanded MolProbity validation with MolProbity score, omega analysis, Rama-Z, bond and angle RMSD, and mmCIF support.

Added files and preview output tabs with filename preservation to client-side converters.

Improved AutoDock Vina receptor preparation, pose validation, and settings.

Added AutoDock-GPU frontend search-space validation.

Added input validation for LigandMPNN homo-oligomer settings and DiffAb chain IDs.

Improved onboarding so returning workspace users are not shown duplicate onboarding.

Improved batch workspace and job history views.

Added raw/format toggle support to FASTA text inputs.

Added upload-first PDB viewer behavior with multi-file support.

Improved structure upload validation and FoldSeek outputs.

Improved SuperWater GPU tiers and credit scaling.

Improved ViennaRNA runtime behavior and result fidelity.

Added RAxML-NG for maximum-likelihood phylogenetic tree inference with bootstrap support.

Added queued job status, safer batch child deletion, and batch workspace queue flow.

Added onboarding for new users.

Added multiple-ligand support to AutoDock Vina single mode for batch submissions.

Added GPU tier selection and credit calculation to SuperWater.

Improved phylogenetic tree outputs with stat cards, file downloads, and collapsible Newick display.

Improved IQ-TREE, Boltz-2, ADMET-AI, ESMFold, and MPNN input validation and execution.

Reorganized the tool directory into clearer analysis, alignment, screening, visualization, and related categories.

Improved batch child resolution and pricing accuracy.

Improved pricing card behavior and billing-period handling.

Improved job delete and rename behavior.

Improved FASTA parsing consistency across tools.

Improved docs category and callout formatting.

Improved account permission and billing/docs loading performance.

Improved PDB result sorting stability.

Added GPU tiering and residue-based credit calculation to DynamicBind.

Improved BioPhi result rendering.

Added PDB validation for misaligned residue-name columns in CHARMM-GUI glycan outputs.

Refreshed popular and new-tool lists on the app dashboard.

Improved docking tools to use native file formats more directly across SMINA, GNINA, AutoDock Vina, and related ligand workflows.

Improved BindCraft, GROMACS, PDB Fixer, and ligand-tool behavior.

Added SMILES valence warnings for over-bonded atoms.

Added paired-molecule validation to PROPKA 3.

Updated ToxPred 2.0, Chou-Fasman, and GNINA tool definitions and docs.

Improved tool settings validation, search inputs, dropdowns, and tab styling.

Improved SMINA, GROMACS, AutoDock Vina, and BindCraft execution and result behavior.

Added batch runs for screening ligands against multiple receptors with AutoDock Vina.

Improved LigandMPNN settings and output capture.

Added flexible-residue validation for AutoDock Vina.

Improved MSA tool behavior.

Improved AI assistant follow-up questions and message submission reliability.

Restored invoices in the billing panel.

Improved SurfDock and SigmaDock execution and result behavior.

Added SurfDock surface-informed diffusion docking for protein-ligand complex prediction.

Improved signup redirects, header credit display, and welcome email styling.

Updated the AI assistant model for stronger bioinformatics guidance.

Added SigmaDock fragment-based SE(3) diffusion docking for protein-ligand binding poses.

Added SigmaDock documentation and promotional pricing.

Removed the SigmaDock residue limit.

Improved chart SVG export branding and theme switching.

Improved TXT to FASTA file preview support.

Improved the AI assistant, docking page experience, and tool UI quality-of-life behavior.

Improved tool reliability and user interface behavior across multiple tools.

Added Ligand fixer for repairing SDF, MOL, and MOL2 ligands for docking preparation workflows.

Added SMILES to SDF, SMILES to PDB, SMILES to MOL2, and SMILES to InChI converters.

Added SDF to SMILES, MOL2 to SMILES, and InChI to SMILES converters.

Improved multiple tools and the shared user interface.

Improved Boltz-2, DiffDock-L, HMMER, and HADDOCK3 behavior.

Improved HADDOCK3 execution and documentation.

Improved tool UI and reliability across several tools.

Improved IgGM execution behavior.

Refreshed the app interface and resolved emergency production issues.

Added batch-mode support to PDB Fixer.

Improved app and tool interface behavior.

Added Salmon for transcript-level RNA-seq quantification with selective alignment.

Added SuperWater for diffusion-based hydration site prediction in protein structures.

Added IntelliFold 2 for controllable structure prediction across proteins, ligands, DNA, RNA, and complexes.

Added RFantibody for antibody and nanobody binder design.

Improved job reliability.

Improved payment flows, Boltz-2, AutoDock Vina, ipSAE outputs, and product documentation.

Added ProFam for family-conditioned protein sequence generation from FASTA or MSA input.

Updated ADMET-AI to a newer version.

Improved structure output performance and multi-file download reliability with server-streamed ZIP exports.

Improved the structure viewer and restored client-side tool behavior.

Improved billing upgrades and downgrades.

Adjusted Boltz-2 pricing.

Improved ADMET-AI, Boltz-2, pricing, and billing behavior.

Added clearer RFDiffusion input instructions.

Improved old job history display.

Improved team and payment workflows.

Updated pricing.

Improved pricing behavior and AutoDock Vina ligand fragment handling.

Prepared pricing updates and related tool quality-of-life improvements.

Improved the shared server-tool experience and visual palette.

Improved the job interface and spreadsheet result display.

Centralized app routing without changing user-facing URLs.

Improved team switching.

Improved AutoDock Vina quality-of-life behavior.

Introduced the next-generation tools layout.

Improved AutoDock Vina and PDB Fixer behavior.

Improved team and billing workflows.

Improved team settings and account workflows.

Improved AutoDock Vina setup and results.

Added AllMetal3D for predicting metal and water binding sites in protein structures.

Added DeepEMhancer for sharpening and denoising cryo-EM maps.

Added DR-BERT for predicting intrinsically disordered protein regions and residue-level disorder probabilities.

Added GROMACS for classical molecular dynamics simulations.

Improved login, menu, and app performance behavior.

Improved workflows and general tool quality-of-life behavior.

Added Protein charge plot for plotting net charge versus pH.

Added FindPept for matching experimental masses to theoretical digest fragments.

Added Glycosylation site finder for identifying N-linked glycosylation sequons.

Added Protein motif scanner for scanning biologically important protein motifs.

Added Protein scale profiler for plotting amino acid property profiles across 42 scales.

Added Peptide cutter for mapping protease and chemical cleavage sites across protein sequences.

Added Peptide mass calculator for in-silico digestion and peptide mass calculation.

Improved Stripe checkout and billing reliability.

Improved settings and Stripe checkout workflows.

Added working ipSAE outputs for structure-confidence analysis.

Improved tool and app behavior across several updates.

Added DeepImmuno deep learning-based peptide immunogenicity prediction against 20 supported MHC-I alleles, with single-HLA and per-sequence HLA assignment modes.

Added TLimmuno2 transfer learning-based MHC-II immunogenicity prediction for CD4+ T cell epitope analysis.

Added Humatch antibody humanization with optimal V-gene matching.

Improved AutoDock Vina receptor preparation, flexible residue handling, and search box sizing for better docking accuracy and performance.

Improved Protenix reliability for jobs that were failing to process correctly.

Fixed AutoDock Vina crashing when users accidentally uploaded protein-sized molecules as ligands by adding atom count validation.

Added AlphaGenome variant effect predictor for analyzing how genetic variants impact gene expression across up to 1M base pair regions.

Added EquiDock SE(3)-equivariant graph neural network for deterministic rigid protein-protein docking.

Added HighFold cyclic peptide structure prediction using modified AlphaFold2 with CycPOEM encoding.

Added PandaDock physics-based molecular docking with sub-angstrom accuracy without GPU requirements.

Added ImmuneBuilder structure prediction for antibodies, nanobodies, and T-cell receptors, 100x faster than AlphaFold2.

Added DiffAb AI-powered antibody CDR design using equivariant diffusion models for antigen-specific optimization.

Added AF-Cluster for predicting alternative protein conformations by clustering MSAs with DBSCAN.

Added ORF Finder to identify open reading frames across all six reading frames with multiple genetic code support.

Added CpG Island Finder for identifying methylation-related regions in DNA sequences.

Added DNA Shuffle for generating randomized control sequences while preserving k-mer composition.

Added Filter DNA and Filter Protein sequence cleaning utilities for removing non-standard characters.

Added GenBank Feature Extractor for extracting CDS, mRNA, and other features from GenBank files.

Added Chain Fetcher input option for fetching specific chains from PDB structures by ID.

Added ViennaRNA suite with 14 RNA analysis tools including RNAfold, RNAalifold, RNAcofold, RNAplfold, and more for secondary structure prediction and thermodynamic analysis.

Added IgGM generative foundation model for antibody and nanobody design with CDR design, affinity maturation, and structure prediction capabilities.

Added IgDesign antibody CDR sequence design via inverse folding for therapeutic antibody development.

Added AntiFold inverse folding tool for antibody variable domain structures using IMGT numbering.

Added Admetica ADMET property prediction using Chemprop-based machine learning models for 22 pharmacokinetic and toxicity endpoints.

Added gmx_MMPBSA binding free energy calculator using MM/PBSA and MM/GBSA methods for protein-ligand, protein-protein, and protein-DNA complexes.

Added AI Assistant chat interface for getting bioinformatics guidance, tool recommendations, and help interpreting results.

Fixed ProstT5 tool that was failing to process sequences correctly.

Added MD Trajectory Analysis tool using MDAnalysis to calculate RMSD, RMSF, radius of gyration, hydrogen bonds, PCA, and 20+ other analyses from MD simulation trajectories.

Added DLKcat deep learning model for predicting enzyme turnover numbers (kcat values) from protein sequences and substrate structures.

Added ODesign all-atom generative AI for designing protein binders with precise control over target binding sites and interaction parameters.

Added MiniFold ultra-fast single-sequence protein structure prediction, 10-20x faster than ESMFold with no MSA required.

Added EvoPro genetic algorithm-based protein binder optimization using AlphaFold2 fitness evaluation and ProteinMPNN sequence design.

Added CleaveNet AI-powered prediction of matrix metalloproteinase (MMP) cleavage sites for peptide sequences.

Added AbLang antibody residue restoration using a language model trained on the Observed Antibody Space database.

Added AbLang-2 antibody-specific language model for predicting non-germline residues with germline bias correction.

Added PAINS Filter to screen compounds for pan-assay interference patterns that cause false positives in biological assays.

Added Brenk Filter to identify toxic, reactive, and pharmacokinetically problematic molecular fragments.

Added Veber's Rule calculator to predict oral bioavailability by evaluating molecular descriptors.

Added Lead-likeness Filter for stricter compound screening in early-stage drug discovery.

Added new Security page at /security outlining platform security practices.

Added fpocket protein pocket detection tool using Voronoi tessellation to identify binding sites with druggability scores, volume, and polarity analysis.

Added Boltz-2 guardrails for protein structures that exceed 1,400 residues.

Fixed PDB2PQR accepting small molecule PDB files and failing with unclear errors.

Added PepMLM peptide masked language model for generating sequence embeddings and predicting peptide properties.

Added SmrTNet deep learning model for predicting small molecule retention times in liquid chromatography.

Added ThermoMPNN protein thermostability prediction tool using message passing neural networks to estimate melting temperatures.

Fixed MolProbity and PDB2PQR failing on PDB files starting with CRYST1 records instead of HEADER/ATOM.

Added IgBLAST immunoglobulin and T cell receptor sequence analysis tool for identifying V/D/J gene segments and CDR regions.

Added mBER VHH nanobody binder design tool using AlphaFold-Multimer to generate novel antibody sequences.

Added DockQ docking quality assessment tool for comparing predicted structures against native references with DockQ scores, iRMSD, LRMSD, and fnat metrics.

Added AlphaFlow protein conformational ensemble generator using flow matching to explore protein flexibility and dynamics.

Added eToxPred machine learning-based toxicity and synthetic accessibility prediction for drug candidate screening.

Added AutoDock-GPU GPU-accelerated molecular docking using the AutoDock4 force field, up to 56x faster than CPU docking.

New General Settings page at Settings → General for managing profile information, work type, active sessions, and account settings.

Enhanced Hydrophobicity Plot and Hydropathy Plot with tabbed Chart/Data views and improved export options.

Improved API performance with signed URL caching for faster file downloads.

Fixed RFdiffusion3 failing on PDB structures with missing residues (gaps in numbering).

Fixed AlphaFold2 failing due to model weights not loading correctly.

Added IQ-TREE phylogenetic tree builder with maximum likelihood and automatic model selection (ModelFinder).

Added MUSCLE5 multiple sequence alignment using the PPP algorithm for high-quality alignments.

Added HMMER for sensitive homology search using profile hidden Markov models to detect remote homologs.

Added HADDOCK3 integrative protein-protein docking platform guided by experimental data.

Added MSA Viewer interactive visualization for multiple sequence alignments with color schemes.

Added ProstT5 protein language model for bidirectional translation between sequences and 3Di structural tokens.

Added full-screen mode for tool outputs with expanded visualization controls.

Added batch submission functionality for submitting multiple jobs at once with progress tracking.

Added DSSP secondary structure assignment tool using hydrogen bond-based analysis (gold standard method).

Added SASA Calculator for calculating solvent accessible surface area using the Shrake-Rupley algorithm.

Added RMSD Calculator to compare protein structures with automatic Kabsch alignment.

Added GenMol NVIDIA generative AI model for creating novel drug-like molecules using masked discrete diffusion.

Added DynamicBind AI-powered protein-ligand binding prediction with conformational change modeling.

Added BindCraft de novo protein binder design using AlphaFold2 backpropagation and ProteinMPNN.

Added Smart Actions dropdown to tool outputs for quick access to common actions like sending results to another tool.

Added ESM-2 protein language model for generating sequence embeddings useful for machine learning pipelines.

Added ESM-IF1 inverse folding tool to design protein sequences from 3D backbone structures.

Added RFdiffusion3 all-atom generative diffusion model for designing protein binders, enzymes, and symmetric assemblies.

Added RosettaFold3 open-source structure prediction for proteins, nucleic acids, and small molecule ligands.

Added Hydrophobicity Plot visualization tool for analyzing protein hydrophobicity profiles using multiple scales.

Added MolProbity structure validation tool with clashscore, Ramachandran, and rotamer analysis.

Added PROPKA 3 for predicting pKa values of ionizable residues in protein structures.

Added AlphaFold2 structure prediction via ColabFold with MSA-based folding for proteins and complexes.

Added MAFFT multiple sequence alignment tool for accurate alignment of protein and nucleotide sequences.

Added FoldSeek structure search tool for finding structurally similar proteins in databases.

Added Radius of Gyration calculator for measuring protein compactness from PDB structures.

Added Clustal Omega multiple sequence alignment tool for large-scale protein sequence alignments.

Added FastTree phylogenetic tree builder for rapid construction of evolutionary trees from alignments.

Added MMseqs2 ultra-fast sequence searching and clustering tool.

Added USalign structure alignment tool for comparing and superimposing protein structures.

Added ParaSurf binding surface prediction tool for identifying protein-protein interaction sites.

Added PDB2PQR for preparing structures for electrostatics calculations with PROPKA integration.

Added Protenix open-source AlphaFold 3 implementation for predicting protein, RNA, DNA, and ligand complexes.

Added CSV to FASTA converter for transforming tabular sequence data into FASTA format.

Added ESMFold protein structure prediction using Meta's evolutionary scale modeling.

Added OpenFold 3 open-source structure prediction model.

Added PocketFlow binding pocket detection and analysis tool.

Added Hydropathy Plot visualization tool for analyzing protein hydrophobicity profiles.

New dashboard recent jobs and quick actions.

Added SVG export functionality for chart visualizations.

Added SMINA molecular docking tool with improved scoring functions for virtual screening.

Added ColabDock protein-protein docking tool with distance restraint support for guided docking predictions.

Added Chai-1 state-of-the-art biomolecular structure prediction model supporting proteins, nucleic acids, and small molecules.

Added RFDiffusion 2 next-generation protein backbone design with improved unconditional generation.

Added Ramachandran Plot visualization tool for analyzing protein backbone dihedral angles.

Added six new protein property calculators: Aliphatic Index, Extinction Coefficient, GRAVY, Instability Index, isoelectric point, and Molecular Weight.

Added job sharing functionality - share your analysis results with collaborators via shareable links.

Added DNA to RNA and RNA to DNA sequence conversion tools.

Added CIF format support for BoltzGen structure inputs.

Added Boltz-2 biomolecular foundation model for structure and binding affinity prediction. It's currently in early beta.

We made four of our premium tools now free: Protein parameters, Amino acid composition, Chou-Fasman, and protein stability. Guests can submit up to 10 FASTA sequences at a time, while logged in users (even on the Free account) can submit unlimited number of sequences. Previously, these tools cost 1 credit per FASTA.

Added LightDock protein-protein docking tool using Glowworm Swarm Optimization for predicting macromolecular binding modes.

Added citation functionality - tools now include citation snippets for academic use.

Added documentation for Text to FASTA converter with usage examples.

Enhanced molecular docking visualization with improved 3D rendering controls.

ProteinIQ now automatically generates citations for the tools you use. You can find them button in the bottom right part of the header. We'll be adding support for this feature to all the tools soon.

Added RFDiffusion protein structure design tool for generating novel protein backbones with diffusion-based modeling.

Enhanced legal documentation: added Computational Usage Policy, Content Policy, DMCA, and Data Processing Agreement.

Improved DNA to Protein translator with better validation and documentation.

Enhanced documentation section with new guides on Credits, Jobs, Pricing, and Start Guide.

Improved credit calculation system with tool-specific calculators including atom-count-based pricing for BoltzGen.

New Usage Dashboard at Settings → Usage showing credit consumption history, job statistics, and usage analytics with interactive charts.

New Billing Management at Settings → Billing for managing subscriptions, viewing invoices, and updating payment methods.

Stripe integration improvements for smoother checkout, subscription management, and payment processing.

Fixed SMILES validation to properly handle edge cases.

Added one of the best models of 2025: BoltzGen. BoltzGen is an expensive model to run and thus requires at least 200 credits, depending on the parameters chosen.

Added three new tools from the same model family: LigandMPNN, ProteinMPNN, and SolubleMPNN.

Added GNINA molecular docking tool with advanced CNN-based scoring and energy minimization.

Improved 3D molecular docking visualization with enhanced ligand-protein rendering and interactive controls.

Enhanced job polling system for more reliable real-time result updates.

Added Autodock Vina molecular docking tool with flexible ligand-protein binding predictions.

Added DiffDock-L AI-based molecular docking model.

Added the Guides section.

Major improvements to how you input sequences and files into tools, both free and paid tools.

Rebuilt the Protein stability prediction tool from the ground up. It's now more solidly based on peer-reviewed methods.

Added real-time input validation for SMILES, FASTA, AND PDB inputs, catching invalid formats before job submission. This means 90% fewer job errors.

Added the FASTA to FASTQ converter which assigns mock quality scores to FASTA sequences.

All tools now share the same improved interface with horizontal input and settings on the left and output on the right. You can simply wait for the results of your job on the same page or you can come back later at any time and access your past jobs via the "History" in the output panel. All tools share the same layout, which makes it easier for you to pick and start using new tools immediately.

Improved and added new settings to the One to three converter.

PDB Fixer now supports solvent box and lipid membrane addition for molecular dynamics simulations, with configurable ion types and box geometries. Performance improved from several minutes to ~40 seconds. We also updated the pricing for PDB Fixer, reduced from 5 credits flat to 2 credits base + multiplier based on atom count.

ProteinIQ now has a dark mode. You can switch between Light, Dark, and System modes by clicking your user icon in the bottom left or through the footer.

DNA mutator has improved color coding and now works as expected.

The first set of cloud compute tools now generally available, including protein parameters calculator, ADMET-AI, and others. Browse all tools.

Added FASTA splitter utility tool.

More tools, including server-side compute now available.

ProteinIQ launches with 13 browser-based molecular analysis tools covering protein parameters, drug discovery (Lipinski, toxicity prediction), sequence processing (FASTA/FASTQ conversion, translation), and structure visualization.

No installations or expensive licenses required. Cloud infrastructure scales automatically with batch processing support and standard format exports.

Roadmap includes AlphaFold integration, molecular dynamics, protein design tools, and API access for pipeline integration.