Skip to content

What is bindsight, and why does it matter?

A 5-minute read for anyone — biologist, data scientist, PI, investor, or recruiter — who wants to understand what this tool is, what it does, and why it's a genuine deal-breaker rather than yet another wrapper.

The one-sentence pitch

bindsight is the first open-source tool that takes RNA-seq counts as input and produces ranked, structurally-validated de novo protein binder candidates as output, with a complete audit trail back to the patient cohort the targets came from.

That's it. Counts in, designed binders out, every step documented.

The two worlds it bridges

In computational biology in 2026, two ecosystems run side-by-side and barely talk to each other:

World Tooling Where it stops
Genomics DESeq2, edgeR, Seurat, scanpy, TCGA, recount3, GTEx "Here are the differentially-expressed genes."
Protein design RFdiffusion, ProteinMPNN, BindCraft, BoltzGen, AlphaFold, Boltz-2, Chai-1 "Given this target structure, here are some designed binders."

Between them sits a chasm. Going from "this gene is interesting" to "here's a protein I designed to bind it" requires:

  1. Mapping gene IDs → UniProt accessions
  2. Filtering for surface-exposed proteins (else you can't drug them)
  3. Filtering for tissue specificity (else you'll hit healthy organs)
  4. Pulling structures from AlphaFoldDB or PDB
  5. Identifying druggable epitopes on those structures
  6. Running a backbone diffusion model (RFdiffusion, etc.) on each
  7. Designing sequences with ProteinMPNN
  8. Validating each design with Boltz-2 / AlphaFold
  9. Multi-objective ranking
  10. Tracking provenance so a reviewer can audit the chain

Today, doing all of this for one disease takes a competent grad student 4–6 weeks of glue scripting, four different conda environments, and at least one HPC allocation. The work is rarely reproducible across labs. The Apr 2026 bioRxiv preprint that did it for one cancer (DSRCT) needed a custom agent framework just to keep state.

bindsight does it in a single command, on a CPU laptop, with reproducibility that survives review.

Why this is a deal-breaker, not yet-another-wrapper

1. It closes a structural gap, not a polish gap

There is no other open-source tool that takes RNA-seq counts as its input type. ProteinDJ, Ovo, BindCraft, dl_binder_design, nf-binder-design, and Tamarind.bio all start with "give us a target structure." They polish the protein-design half of the pipeline. The discovery half — the genomics-to-structure bridge — is where we live, and where nobody else has shipped.

2. The keystone (SURFACE-Bind) just dropped

The SURFACE-Bind catalog (PNAS 2025) ships pre-computed targetable interfaces + binder seeds for ~2,800 surface proteins under a BSD-3 license. Before this, "find a druggable site on an arbitrary surface protein" was its own project. Now it's a UniProt-keyed lookup. This unlocks the bridge for a one-person team — exactly the right moment to build.

3. CPU laptop is enough

The orchestrator runs on the user's machine. GPU work runs end-to-end on free Colab T4s, free Kaggle T4×2s, paid Modal A100s, or a local NVIDIA GPU — the user picks the backend per command. The user's $0/month research budget is no longer the bottleneck; the same setup that works for a PhD student in Cairo works for a clinician in Lagos.

4. Provenance is the moat

Every binder PDB the pipeline outputs is one click from:

  • the patient cohort it came from,
  • the differential-expression statistics that flagged the target,
  • the AlphaFoldDB structure used,
  • the SURFACE-Bind site predicted,
  • the trajectory seed and designer commit SHA,
  • the validator metrics,
  • the container digest of every step.

This is recorded as a PROV-O JSON-LD manifest and packaged as an RO-Crate — both W3C/FAIR-friendly standards. Reviewers can audit the chain. Clinicians can defend the choice in IND filings. No other protein-design tool offers this.

5. Failure-honest by default

The pipeline catalogs why targets fail (no AlphaFoldDB model, no SURFACE-Bind site, fails specificity, designer fails to converge, validator rejects), publishing a failure_taxonomy.parquet next to the successes. Every existing tool quietly drops failures. We surface them, because that's what users actually need to triage.

6. Commercially defensible

The default config uses only MIT / Apache / BSD / CC-BY components. Every non-permissive opt-in (e.g. AlphaFold2 weights for the AF2-IG validator) is behind a CLI banner and documented in LICENSING.md. A pharma early-discovery team can run it without legal review.

What you can do with it

Discovery half — CPU, no GPU

bindsight discover examples/tcga_luad.yaml --out runs/luad_v01

Produces:

  • runs/luad_v01/deg/results.parquet — DEG table from pydeseq2
  • runs/luad_v01/targets/candidates.parquet — surface-restricted, druggable, tissue-specific target shortlist
  • runs/luad_v01/epitopes/epitopes.parquet — top-N candidates with structure paths
  • runs/luad_v01/run_manifest.jsonld — PROV-O audit trail

Useful right now if you're a translational researcher trying to triage a cancer cohort for surface-antigen targets without writing 600 lines of glue.

Design + validation half — GPU-backed

bindsight design runs/luad_v01 --backend colab --trajectories 50
bindsight validate runs/luad_v01 --backend colab --validator boltz2
bindsight rank runs/luad_v01
bindsight report runs/luad_v01 --format html

Adds:

  • Per-target de novo binder PDBs from RFdiffusion + ProteinMPNN
  • Boltz-2 affinity + iPTM scores per design
  • Composite ranking
  • Self-contained HTML report (embedded volcano plot + tables + PROV-O manifest)
  • RO-Crate export for Zenodo deposit

Validation (done — discovery half)

A companion report runs the discovery half on six real indication-matched TCGA cohorts. It rediscovers ERBB2 at rank 4 in HER2-enriched breast cancer (using PAM50 subtype stratification — versus rank 25 in the unsplit BRCA cohort, where the HER2 signal is averaged away) and is specific: antigens that are not transcriptionally over-expressed at the bulk level (EGFR, which is mutation-driven; CEA, co-expressed in normal colon) are correctly not surfaced. Reproducible artifacts are in benchmarks/validation/. The three-way designer benchmark is GPU-only; a runnable, CPU-tested harness + protocol ship in benchmarks/designer_benchmark/.

v1.0

Multi-modal tumor-selectivity scoring (single-cell + co-expression + immunopeptidomics) to extend discovery beyond bulk differential expression, plus the populated designer benchmark.

Who this is for

Audience What they get
Translational researcher with a TCGA cohort and limited compute A reproducible "data → designed binder" pipeline without paying SaaS prices
Clinical biologist who needs an audit trail A PROV-O / RO-Crate trace from binder back to expression evidence — defensible at thesis defense and useful for IND filings
Method developer building a new designer or validator A held-out evaluation harness (rediscovery of known antigens) to benchmark against a fixed upstream pipeline
Pharma early discovery team A free open-source comparator they can layer their proprietary designers into via the plugin interface
Educator teaching computational biology A working end-to-end example that touches DEG, structural biology, deep learning, and software engineering — in one repo
PI hiring for a bioinformatics + ML role A CV artifact that demonstrates real cross-domain competence

What it is NOT

  • Not a wet-lab pipeline. Output is in silico binder candidates. Wet validation (yeast display, biolayer interferometry, etc.) is the user's job.
  • Not a regulated GxP tool. The provenance graph is a starting point for an audit trail, not a substitute for a GxP-validated system.
  • Not a structure predictor itself. It wraps Boltz-2 / Chai-1 / AlphaFold; the predictions are theirs, attributed in every manifest.
  • Not a replacement for domain expertise. A surface-protein hit list still needs a human to apply biological judgment about safety, specificity, immunogenicity, and developability.

The honest limits in v0.x

We're transparent about what doesn't work yet. From ARCHITECTURE.md § 10:

  • pydeseq2 is not bit-equivalent to R DESeq2 (documented)
  • SURFACE-Bind covers ~2,800 surface proteins, not all of them (graceful drop)
  • Free Colab sessions die mid-job (per-trajectory checkpointing mitigates)
  • Designer choice (RFdiff vs BindCraft vs BoltzGen) will keep evolving (plugin interface; the rfdiff_mpnn arm is benchmarked today — see the designer benchmark — with BindCraft/BoltzGen on paid ≥24–32 GB GPUs as it grows)
  • mRNA abundance is not cell-surface protein abundance — SURFY confirms a protein can reach the surface, not how much is actually there. Surfaced candidates are hypotheses to confirm at the protein level (flow cytometry / IHC / Human Protein Atlas) before trusting them. (Stated in every run's report Limitations section.)
  • Bulk expression can come from non-tumour cells — high apparent over-expression may reflect infiltrating immune/stromal cells or tumour purity rather than a tumour-intrinsic target; single-cell / deconvolution evidence is needed to be sure (the multi-modal tumour-selectivity layer for v1.0).
  • Disease specificity is hard — "up in cancer, low in vital tissue" predictably finds known antigens (a feature for the v0.1 rediscovery paper, a problem layer for v1.0)

Why this is the right project at the right time

  • The OSS protein-design stack just consolidated. Boltz-2 (MIT, code + weights), BoltzGen (MIT), Chai-1r (Apache-2), and BindCraft (MIT) all shipped permissive licenses in 2025. A year earlier, the "what validator do we ship?" question had no clean answer.
  • SURFACE-Bind closed the epitope-prediction gap for a meaningful slice of the surfaceome. Before late 2025, you'd have built that yourself.
  • Free GPU tiers (Colab T4, Kaggle T4×2) became powerful enough to run RFdiffusion + ProteinMPNN at meaningful scale. A year earlier, you'd have needed an institutional cluster.
  • Cancer omics is publicly available (TCGA, recount3, GTEx) — the upstream data is already in the open.
  • The clinical translation gap is widely acknowledged (e.g. Springer J Transl Med 2026). Reviewers want to see the bridge built.

All four conditions — permissive validators, the SURFACE-Bind catalog, free GPU, and acknowledged demand — held simultaneously starting in late 2025. That's the window.

How is this different from "just calling AlphaFold + RFdiffusion"?

Anyone with a GPU can run the existing tools. The work bindsight does is the opinionated, validated, reproducible join. Specifically:

  1. Empirical defense of the defaults. The discovery half is validated by rediscovery on six real TCGA cohorts (benchmarks/validation/): it surfaces ERBB2 at rank 4 in HER2-enriched breast and is specific against antigens that aren't transcriptionally over-expressed. See the validation report.
  2. Container-pinned, seed-pinned, weights-pinned reproducibility. Two runs of the same config on the same data should produce byte-identical outputs modulo logged stochastic seeds.
  3. Negative-result curation. A failure_taxonomy.parquet per run — every target that didn't make it, with a reason.
  4. Cost-aware orchestration. --dry-run estimates GPU $ before running. Existing tools assume HPC and don't think about cost.
  5. Plugin-not-fork extensibility. New designers / validators / runners ship as separate Python packages that register via entry points. Nobody has to fork to extend.

Summary

Two ecosystems that should be one. A keystone catalog (SURFACE-Bind) that just shipped. Permissive validator licenses that just landed. Free GPU tiers that just got good enough. Public cancer omics. A reproducibility crisis everyone agrees on. One person can connect all of this for the first time. That's bindsight.

If you want the architectural detail, see ARCHITECTURE.md. If you want to use it now, see how-to-use.md. If you want to see what it can do, see use-cases.md.