--- profile: elgora_markdown_bounty_challenge_v0 escrow_amount: "1000000" submission_deadline: 1789120800 payout_policy: winner_take_all --- # Adaptyv RFdiffusion IL-7Rα binder BLI revalidation ## Summary Submit one Baker-lab IL-7Rα miniprotein from Adaptyv's RFdiffusion revalidation collection. Guardians score published BLI `kd`. They do not run BLI. ## Challenge details Adaptyv retested RFdiffusion-designed IL-7Rα miniproteins (Cao et al. / Baker lab designs) with a standardized BLI affinity workflow and published sequences plus kinetic fields on Proteinbase collection `rfdiffusion-re-validation`. This is a wet-lab methods revalidation, not the EGFR or TREM2 community contests. Forty-two proteins are in the hashed snapshot; 15 have experimental `kd`. The `author` field is empty on these rows. This bounty purchases one competitive published miniprotein from that retest. It does not purchase a new wet-lab campaign. ### Definitions And Scope A candidate is the amino acid string in `binder.fasta`. A match is the unique row with that `sequence`. Success establishes a published expressed IL-7Rα binder with a usable experimental `kd`. It does not establish a new physical sample or that the Solver performed the Baker-lab design protocol. ## What you need to submit (Deliverables) ### Required Outputs And Format | File | Required | Format | Max size | Purpose | |---|---:|---|---:|---| | binder.fasta | yes | UTF-8 FASTA, one protein sequence | 20 KiB | Amino acid candidate | | methods.md | yes | UTF-8 Markdown | 100 KiB | Disclose `name` and `designMethod` | binder.fasta: one `>` header, then amino acids. Ignore the header. Strip ASCII whitespace. Nonempty. Only `ACDEFGHIKLMNPQRSTVWY`. methods.md must contain the matched row's `name` and `designMethod` as case-sensitive contiguous substrings. Do not require `author`; it is empty in this snapshot. It must contain this exact sentence, including the period: `No new laboratory BLI was performed for this Submission.` Package rules: - archive format: none; one flat directory; - only `binder.fasta` and `methods.md`. ## Input Files References | File | Why it is needed | How to get it | SHA-256 content hash | |---|---|---|---| | il7ra_revalidation.csv | Proteinbase snapshot of 42 RFdiffusion IL-7Rα retest designs with BLI fields | Public HTTPS GET, no login: https://proteinbase.com/api/proteins/download?collectionId=0199a4d3-350c-5581-110e-3128a190f1c8&slug=rfdiffusion-re-validation | `fe8d6235e4a9e7ca567397bd88ebbae0e2e0a98673dc341cca361a6859e83029` | ### Access And Known Limitations Guardians fetch that URL themselves and check SHA-256 of the raw bytes, including a UTF-8 BOM if present, with no decode before hashing. Parse with UTF-8-SIG. `evaluations` is JSON and includes `bli_kinetic_curves`. Do not fetch curve URLs. Missing access or a hash mismatch blocks judgment. There is no Proteinbase signature, notarization, or signed manifest for this snapshot. Guardians cannot independently prove Proteinbase authored the bytes. This bounty purchases historical analysis of the exact bytes whose SHA-256 is listed above. ## Acceptance Criteria ### Pass/Fail Checks 1. Exactly two files, `binder.fasta` and `methods.md`. 2. Size limits 20,480 and 102,400 bytes. 3. FASTA alphabet rule. 4. Exactly one matching `sequence` row. 5. That row's `designMethod` equals the case-sensitive exact string `rfdiffusion`. Do not lowercase. `RFdiffusion` is not that value. 6. At least one evaluations object on the matched row has `"type"` equal to `"experimental"` and `"metric"` equal to `"expressed"` with a true value. Treat JSON `true`, boolean true, and the strings `true` and `True` as true. 7. At least one evaluations object on the same row has `"type"` equal to `"experimental"`, `"metric"` equal to `"binding"`, and a true value using the same true test. `expressed`, `binding`, and `kd` need not be the same object. 8. At least one qualifying experimental `kd` as defined in Scoring. A non-qualifying `kd` object does not satisfy this check. 9. methods.md contains `name`, `designMethod`, and the BLI disclosure sentence. ### Scoring And Calculations Walk the matched row's `evaluations` JSON array from the first element to the last. A qualifying `kd` is an object whose `"type"` equals `"experimental"`, `"metric"` equals `"kd"`, `"target"` equals `"il7r"`, `"unit"` equals `"M"`, and `"valueType"` equals `"numeric"`, all as case-sensitive exact strings, and whose `value` parses with Python 3 `float()` as a finite number strictly greater than `0.0` and at most `1.0`. Units are molar. Do not convert nM or other units. The snapshot has no failed-control field; do not invent one. All qualifying `kd` objects on that one hashed row are treated as comparable BLI measurements because they share this snapshot, target `il7r`, and unit `M`. Collect them in encounter order. Do not sort. Score is: `math.exp(sum(math.log(kd_i) for kd_i in kds) / len(kds))` in IEEE-754 binary64 using Python 3 `math.log` and `math.exp` on that list, left to right. Do not round except as those binary64 operations. If the qualifying set is empty, or `math.exp` or `math.log` raises `OverflowError` or `ValueError`, the Submission is invalid. Lower is better. Do not rank by `esmfold_plddt` or TM-score fields. Binary64-equal scores are ties and use the winner rule below. ### Missing, Invalid, And Conflicting Results - Empty `author` is expected and is not a failure. - No experimental `kd`: invalid. - Unavailable fetch is an operational blocker. ### Evidence And Provenance The Poster selects the Proteinbase collection `rfdiffusion-re-validation` at the listed URL as the source of this historical BLI table. File identity is the SHA-256 on this page. Guardians fetch the bytes and check the hash. That verifies the selected snapshot, not a new assay and not Proteinbase authorship beyond those bytes. Link a candidate by exact `sequence`. There is no separate producer authentication commitment. ## How is the winner selected? - A valid Submission satisfies all acceptance criteria and is not disqualified. - Lowest geometric-mean `kd` wins. Binary64 ties go to the lowercase Solver address that sorts first. - If no Submission is valid, the outcome is `no_valid_submission`. ## Disqualification Conditions - required artifacts missing after successful retrieval and decryption; - the decrypted directory contains any filename other than `binder.fasta` and `methods.md`. UTF-8 decode failure, FASTA parse failure, or Markdown that cannot be read as UTF-8 text fails the corresponding Pass/Fail check. There is no separate “corrupt artifact” disqualification. ## Out Of Scope Guardians must not rank by `esmfold_plddt` or TM-score fields. methods.md may mention BLI, RFdiffusion, or computational scores; those mentions do not disqualify and do not change the score. Non-`rfdiffusion` rows fail Pass/Fail Check 5. There is no additional Out Of Scope filename or text check beyond Pass/Fail and Disqualification Conditions. ### Allowed Resources And Reuse Published retest designs may be submitted. Disclose `name` and `designMethod`. ## Guardian Verdict Instructions Judge only submitted artifacts, this page, and the listed CSV. Do not run BLI. ### Evaluation Procedure And Limits 1. Fetch `il7ra_revalidation.csv` and check SHA-256. 2. Open binder.fasta and methods.md. 3. Apply Pass/Fail Checks. 4. Geometric-mean experimental `kd`. 5. Apply the winner rule. One pass over 42 rows is enough.