ProTox-3 limits & performance
Batch sizes, daily quota, throughput, and endpoint-driven cost.
This page helps you plan around ProTox-3's two hard limits — a per-job compound cap and a shared daily query quota — and shows how your endpoint selection drives processing time. Batch size, how many of the 61 endpoints you request, and how you pace jobs against the daily quota matter most for turnaround.
Not available in chat
ProTox-3 is deprecated on the platform and not registered with the orchestrator, so chat cannot call it. Use ADMET-AI for toxicity flags (hERG, Ames, hepatotoxicity) or DeepPurpose for specific endpoint predictions — existing ProTox-3 job results remain viewable.
Batch sizing
The webapp caps a single job at 50 compounds, and shows a warning as you approach it so you can split before submitting.
- Single compound: near-instant for a small endpoint set.
- Small batch (2–45 compounds): processes as one job; a warning appears at 45 to flag daily-quota consumption.
- Maximum per job: 50 compounds. Split larger sets across multiple jobs.
- Empty compound rows are filtered out before submission, so they do not count against the cap.
To profile a library larger than 50 compounds, break it into 50-compound jobs and attach them all to the same project for tracking. See inputs & parameters for how compounds are entered and validated.
Daily quota
ProTox-3 enforces a shared 250 queries per day limit. Plan around it:
- Each compound-endpoint prediction draws on the same daily budget, so requesting the full comprehensive panel on many compounds consumes the quota faster than a single targeted endpoint.
- The 45-compound warning exists specifically because large batches can exhaust the daily allowance in a few jobs.
- If you hit the quota, wait for the daily reset rather than resubmitting immediately.
Endpoint selection drives cost
Processing time and quota use scale with how many endpoints you select, not just compound count.
- "All Models (Comprehensive)" runs all 61 endpoints and is the heaviest option.
- Selecting a focused set (for example, acute toxicity plus hepatotoxicity) completes faster and stretches the daily quota further.
- Predictions across endpoints are parallelized, so a targeted panel returns sooner than the full comprehensive run.
Throughput
- Multiple jobs can run concurrently; queue priority follows your workspace plan.
- Attach jobs to a project so usage and results are grouped for auditing.
- Monitor consumption against the daily quota on the job pages before launching a large campaign.
Performance optimization
- Group structurally similar compounds so you can reuse the same endpoint selection across jobs.
- Use endpoint selection to reduce computation — request only the toxicity categories your decision depends on.
- Schedule comprehensive runs earlier in the day so a quota reset does not interrupt a batch mid-campaign.
Read next
Was this page helpful?