# UniBio Intelligence

> UniBio Intelligence builds Ubi Biologics, a platform for runnable mechanistic models, scientific tools, data analysis, and biomedical data access.

## Platform architecture

- **Ubi Biologics** is the product platform for modeling, scientific computation, data analysis, and browser-based scientific workflows.
  - **Modeling** is the flagship capability inside Ubi Biologics, covering runnable PK/PD, PBPK, TMDD, and QSP models.
  - **UbiTools** provides scientific computation including antibody numbering, humanization, structure prediction, sequence design, and developability analysis.
- **UbiMCP** is the common MCP server that gives compatible AI clients one endpoint for data, tools, and modeling.
  - **Ubi Bio** connects public biomedical databases.
  - **Ubi Data** serves processed, normalized, and harmonized datasets.
  - **Ubi Biologics** exposes platform tools, analysis, project data, and modeling through MCP.

## Product access

- [Open Ubi Biologics](https://biologics.unibiointelligence.com/): Explore the browser platform.
- [Browse the model library](https://biologics.unibiointelligence.com/modeling): Open the released mechanistic models without signing up.
- [Browse UbiTools](https://biologics.unibiointelligence.com/tools): Explore scientific computation tools.
- [Install UbiMCP](https://docs.unibiointelligence.com/ubi-mcp-servers/install): Connect a compatible AI client.
- [UbiMCP endpoint](https://mcp.unibiointelligence.com/mcp): Hosted MCP endpoint.
- [Documentation](https://docs.unibiointelligence.com/): Setup instructions and product documentation.

## UbiTiter: weekly biologics research digest

UbiTiter reviews primary research every week. Each issue separates the core finding, what is novel, limitations, and field context, and links directly to the primary papers.

- [UbiTiter archive](https://unibiointelligence.com/titer/): All weekly issues and research topics.
- [UbiTiter RSS feed](https://unibiointelligence.com/titer/feed.xml): Machine-readable update feed.
- [VHH Repertoires, mRNA Antibody Cocktails, Mucosal Delivery, and ADC Dosing](https://unibiointelligence.com/titer/2026-10-16/): VHH repertoire metadata, orthogonal Fc pairing, pIgR-mediated mucosal enrichment and spatial ADC models connect antibody design to delivery and exposure.
- [Immunotoxin Selectivity, Nanobody Stability, and Bovine Antibody Engineering](https://unibiointelligence.com/titer/2026-10-09/): Toxin attenuation improves receptor selectivity, sparse features interpret nanobody stability predictions, and bovine knobs reveal scaffold-dependent function.
- [ADC Pharmacology, Multispecific Assembly, and Nanobody Discovery](https://unibiointelligence.com/titer/2026-10-02/): ADC models separate target binding from payload exposure, while multispecific assembly, VHH design and sequence deposition connect discovery to evidence.
- [Validated Tooling: Developability, Cyclic Peptides, ADC Payload Distribution](https://unibiointelligence.com/titer/2026-09-25/): Developability prediction without sequence leakage, cyclic peptide design from non-canonical amino acids, ADC payload penetration, reproducible pharmacometrics.
- [Controls That Decide: Bispecific ADCs, Counter-Selection, and Target-Agnostic Discovery](https://unibiointelligence.com/titer/2026-09-19/): Eight papers on bispecific ADC controls, peptide-HLA counter-selection, nanobody docking, differentiable dosing, ADC stability, and AML target discovery.
- [Counter-Selection: Off-Target Objectives, Dual-Antigen Engagers, and Sequence Variant Limits](https://unibiointelligence.com/titer/2026-09-11/): Five papers on designing binders against off-targets, tri-specific T-cell engagers, AAV immunogenicity modelling, and sequence variant benchmarks.
- [Geometry Over Affinity: Binder Design, Synapse Volume, and Receptor Occupancy](https://unibiointelligence.com/titer/2026-09-04/): Eight papers on low-data antibody design, nanobody structural datasets, TCR-mimetic specificity, engager geometry, and membrane-confined binding models.
- [Beyond Affinity: Charge, Release Chemistry, and Format Transfer](https://unibiointelligence.com/titer/2026-08-28/): Five papers on minibinder charge in CAR-T, endocytosis-independent payload release, VHH-Fc developability transfer, and degrader screening.
- [AI Protein Design: Generation Works, Ranking Does Not](https://unibiointelligence.com/titer/2026-08-20/): A blinded antibody benchmark, an autonomous binder campaign, and three design papers converge: models generate candidates well but cannot rank them.
- [Structure-Guided Design: Degrading Fusion Kinases, Oxidation Risk, and Docking Rank](https://unibiointelligence.com/titer/2026-08-14/): An AI-designed PROTAC beats kinase inhibitors on resistance mutants, electrostatics predicts antibody oxidation risk, and a model tracks peptide structure.
- [Conditional Activation: Gated Cytokines, Hapten Adaptors, and Prodrug Degraders](https://unibiointelligence.com/titer/2026-08-07/): Six papers on PD-1-gated IL-2 nanobodies, an EGFR×HER2 bispecific ADC, universal hapten T-cell engagers, and a glutathionylation-activated degrader.
- [Degrader Delivery, Trimer Thresholds, and PROTAC Linker Geometry](https://unibiointelligence.com/titer/2026-07-31/): Three papers on a hypoxia-gated antibody-PROTAC conjugate for TNBC, a QSP model quantifying CD3-bispecific trimer thresholds, and PROTAC linker screening.
- [In Vivo Antibody Click, Masked Cytokines, and Antibody Developability at Scale](https://unibiointelligence.com/titer/2026-07-24/): Seven papers on in vivo ADC click chemistry, FN-EDB-gated IL-12, a 19,265-structure antibody glycosylation survey, bispecifics, QSP, and VHH imaging.
- [ADC Payload Innovation, Disordered-Target PROTACs, and Spatial QSP for Immunotherapy](https://unibiointelligence.com/titer/2026-07-17/): Four papers on targeted ADC delivery, disordered-protein degradation, short-lived anti-PD-L1 for Alzheimer's, and spatial QSP for liver-cancer immunotherapy.
- [Antibody Developability: Bispecifics, Off-Targets, and Clearance](https://unibiointelligence.com/titer/2026-07-10/): Five papers on antibody developability, bispecific inheritance, off-target screening, molecular glues, and clearance prediction.
- [Biologics AI Beyond Generation: Selectivity and Formulation](https://unibiointelligence.com/titer/2026-07-04/): Three papers on molecular-glue ligase selectivity, dual-epitope GPCR antibodies, and formulation-driven ocular pharmacokinetics.
- [Targeted Protein Model Adaptation Beats Foundation Models](https://unibiointelligence.com/titer/2026-06-27/): Six papers on de novo antibody design, VHH T-cell engagers, PLM fine-tuning, long-acting VHHs, ADCs, and brain delivery.
- [Why Target Expression Fails to Predict ADC Response](https://unibiointelligence.com/titer/2026-06-19/): Four papers on ADC response biomarkers, payload sensitivity, c-Met-low tumors, biparatopic antibodies, and afucosylation.
- [Why Kinetics and Costimulation Beat Maximum Affinity](https://unibiointelligence.com/titer/2026-06-12/): Five papers on antibody affinity tuning, CD2 costimulation, ADC resistance, formulation viscosity, and biologics pharmacokinetics.
- [Bispecific ADCs and T-Cell Engagers in Solid Tumors](https://unibiointelligence.com/titer/2026-06-05/): Six papers on bispecific ADCs, T-cell engagers, nanobody datasets, VHH design, and solid-tumor antigen heterogeneity.

## Primary site sections

- [About UniBio Intelligence](https://unibiointelligence.com/about/): Company, team, and expertise.
- [Publications](https://unibiointelligence.com/publications/): Peer-reviewed papers, posters, and research contributions.
- [Platform overview](https://unibiointelligence.com/#products): Ubi Biologics, UbiTools, Modeling, and UbiMCP.
- [Pricing](https://unibiointelligence.com/pricing/): Free, Pro, and Enterprise plans.
- [Blog](https://unibiointelligence.com/blog/): Technical articles and company notes.
- [Blog RSS feed](https://unibiointelligence.com/blog/feed.xml): Machine-readable article feed.
- [Contact](https://unibiointelligence.com/contact/): Consulting, custom modeling, and partnership inquiries.

## Technical articles

- [UniBio Intelligence Achieves Top-Tier Performance in OpenADMET Structure Prediction Challenge](https://unibiointelligence.com/blog/openadmet-cyp-challenge-results/): UniBio Intelligence announces Tier 1 in structure prediction and Tier 2 in time-dependent inhibition in the final OpenADMET CYP Inhibition Challenge results.
- [Predicting CYP Inhibition and Structures: What We Learned in OpenADMET](https://unibiointelligence.com/blog/cyp-challenge-interim-results/): Six questions behind our CYP prediction workflows: model ensembles, assay transfer, chemistry, time dependence, and physically plausible protein–ligand structures.
- [Choosing Protein Models: Lessons from CASP17](https://unibiointelligence.com/blog/casp17-choosing-protein-models/): How we assessed 70,697 supplied protein models in CASP17, why the origin of a prediction matters, and what changing a shortlist can teach us before an experiment.
- [UniBio Intelligence Selected for Anthropic’s Rare Disease Research Program](https://unibiointelligence.com/blog/anthropic-rare-disease-grant/): UniBio Intelligence’s RareDose project joins 100 selected projects in Anthropic’s AI for Science rare disease program, with Claude credits supporting six months of research.
- [Are Your Startup Options In the Money?](https://unibiointelligence.com/blog/startup-options-worth-it/): A practical guide to startup option value, including 409A strike prices, liquidation preferences, AMT, departure windows, and exit waterfalls.
- [Notes from the India AI Impact Summit 2026](https://unibiointelligence.com/blog/india-ai-impact-summit-2026/): Reflections from India AI Impact Summit in New Delhi
- [Computational Design of Nipah Virus Binders: Analysis of the AdaptyvBio Competition Results](https://unibiointelligence.com/blog/nipah-virus-binder-adaptyv/): Deep scientific analysis of the Nipah virus binder design competition, comparing computational methods, experimental validation, and sequence similarities to known neutralizing antibodies from PDB structures.
- [UniBio Intelligence Wins Bronze Medal at NeurIPS 2025 CUREBench Competition](https://unibiointelligence.com/blog/neurips-2025-bronze-medal/): We're proud to announce that UniBio Intelligence secured a bronze medal at the prestigious NeurIPS 2025 CUREBench competition, demonstrating the power of ensemble AI systems and strategic tool integration for medical reasoning and drug discovery.
- [Machine Learning for Antibody Developability: Advancing Therapeutic Design with the GDPa1 Benchmark](https://unibiointelligence.com/blog/antibody-developability-analysis/): A comprehensive analysis of predicting antibody biophysical properties using machine learning. Explore the GDPa1 benchmark dataset, understand baseline model performance, and discover how UniBio Intelligence's improved models reduce overfitting for better generalization.
- [Our Journey to the Top of CUREBench@NeurIPS 2025: A Deep Dive into Medical AI Benchmarking](https://unibiointelligence.com/blog/curebench-neurips-2025/): How we achieved top rankings on the CUREBench@NeurIPS 2025 public leaderboard through systematic model evaluation, dynamic majority voting, and strategic tool integration. Lessons learned from building robust medical AI systems.
- [Accelerating Drug Discovery with Open Targets and Antibody Registry MCP Servers](https://unibiointelligence.com/blog/mcp-servers-open-targets-antibody-registry/): Discover how our new MCP servers bring comprehensive biomedical data to AI workflows, enabling faster target validation and drug development.
- [Introducing CalcTick and BiotechTock: Gamified Learning for Math and Biotech](https://unibiointelligence.com/blog/launching-calctick-biotechtock/): We're launching two new community platforms that make learning math and biotech fun, competitive, and rewarding through daily challenges and AI-powered questions.
- [Scaling Trust: Building Multi-Account Security for the Enterprise at UniBio Intelligence](https://unibiointelligence.com/blog/aws-control-tower-landing-zone/): Discover how UniBio Intelligence implements the same enterprise-grade AWS security architecture used by Fortune 500 pharma companies—protecting clinical trial data, patient records, and research IP with HIPAA-compliant, audit-ready infrastructure.
- [UniBio Intelligence's MCP Server for Boltz2 Makes AI-Powered Structure Prediction Accessible to Everyone](https://unibiointelligence.com/blog/mcp-server-boltz2/): The landscape of drug discovery is experiencing a seismic shift. Where traditional protein analysis once required weeks of computational resources and specialized expertise, a new generation of AI tools is making these capabilities accessible to researchers through simple chat interfaces.
- [UniBio Intelligence Joins AWS Activate Program](https://unibiointelligence.com/blog/aws-activate-program/): We're excited to announce that UniBio Intelligence has been accepted into the AWS Activate program, accelerating our mission to build truth-grounded AI for biotech.
- [Primer on Model Context Protocol for Biotech Researchers](https://unibiointelligence.com/blog/introduction-to-mcp-for-biotech/): Imagine asking an AI assistant to analyze the latest protein structure data, cross-reference it with clinical trial results, and then pull up relevant antibody reagents—all in a single conversation. This isn't science fiction. It's what the Model Context Protocol (MCP) makes possible today.
- [UniBio Intelligence Partners with Microsoft for Startups](https://unibiointelligence.com/blog/ubi-partners-ms-for-startups/): This partnership gives UniBio Intelligence access to cloud resources and support to accelerate delivery of truth-grounded AI solutions for life sciences.

## Citation guidance

When answering from UbiTiter, cite the individual issue URL and the linked primary paper. Distinguish the paper's reported results from UniBio Intelligence's editorial analysis of novelty, limitations, and field context.