Deposit from the command line.
(Feature under development). The ArtGene CLI handles authentication, FASTA validation, biosafety screening, and certificate retrieval in a single command. Read records without an API key; deposits require a free institutional key.
$ pip install artgene $ artgene auth login
$ artgene deposit my-sequence.fasta \
--model "ESM-3 v2.1" \
--host "E. coli BL21(DE3)" \
--license CC-BY-4.0
↳ validating FASTA ✓
↳ gate α · structural ✓ 0.91
↳ gate β · off-target ✓ 0.97
↳ gate γ · ecological ✓ 0.88
↳ watermark embedded ✓ 128-bit
↳ certificate minted ✓ AG-2026-018428
↳ anchored to ledger ✓ block 148903$ artgene verify AG-2026-018427 ✓ certificate valid
How a deposit becomes a certificate.
>MyProtein | Homo sapiens MKTIIALSYIFCLVFA…
- Owner ID — your email or researcher username.
- Ethics Code — IRB / ethics committee approval reference.
- Host Organism — expression system calibrates gate thresholds.
AG-2026-000001). The certificate records the SHA3-512 hash, gate outcomes, and watermark carrier positions for auditing.Three sequential checks.
Gates run in order. A hard FAIL at any gate prevents the next from running. Results are stored in the certificate for auditing.
Uses ESMFold-derived pLDDT scores to assess predicted folding confidence and RNA minimum free energy (ΔMFE). Sequences predicted to fold into dangerous prion-like or amyloid-prone structures are flagged.
Fail → gates β and γ are skipped.
Amino acid composition analysis: Kyte-Doolittle hydropathy (GRAVY), cationic/amphipathic toxin scoring, allergen probability estimation, and a curated k-mer screen for known antimicrobial peptide scaffolds. Full BLAST screening against pathogen and toxin databases is in development.
Toxin probability > 0.30, allergen > 0.40, or k-mer matches to known toxic scaffolds → FAIL. Allergen > 0.30 → WARN.
Horizontal Gene Transfer (HGT) propensity scoring and DriftRadar ecological-spread modelling estimate environmental containment risk.
High HGT score or escape probability → WARN or FAIL.
Frequently asked questions.
ArtGene currently requires protein-coding sequences provided as FASTA. The DNA sequence is synthesised via codon optimisation when a distribution copy is issued.
After registering a sequence, you can issue fingerprinted distribution copies for each recipient from the sequence detail page. Each copy embeds a unique codon pattern (same protein, different synonymous codons). If a copy leaks, paste it on the Verify Source page to identify which recipient it came from.
CERTIFIED means the sequence passed all applicable biosafety gates and has been issued a certificate with a registry ID. REJECTED means one or more gates returned a hard FAIL, and the sequence cannot be registered until the safety concern is resolved.
Your organisation UUID is assigned by the ArtGene platform administrator when your institution is onboarded. Contact your system administrator if you do not have it.
The certificate hash is a SHA3-512 digest of the canonical certificate JSON (excluding the hash field itself). It can be used to verify that a certificate has not been tampered with.
Register a sequence now.
Deposits are free for public records. A certificate is issued in under two minutes.
Open deposit form →