Have Claude take a bioinformatics challenge from paper to report

At a hackathon, Pedro Beltrao’s team used Claude agents to work through 13 bioinformatics prediction problems end to end.

Done with Claude· Advanced· Oct 2026

What they did
During an AI hackathon, Pedro Beltrao’s group gave Claude agents 13 prediction tasks in bioinformatics. For each one, the agents moved through the full sequence on their own: they reviewed the published work, picked a method, wrote the program, ran tests and benchmarks, and assembled a write-up.
How it went
The group had outputs for 11 of the 13 tasks, with 2 still running when the overview was posted. The results had not yet been checked much by humans, so their reliability remained unconfirmed.
Tips
Use this kind of workflow when the problem, inputs, and success criteria are already well defined. Plan on a manual review step after the agent finishes, especially for code and scientific conclusions.

Try it yourself with Claude

I have a [bioinformatics prediction challenge] with [description and data link]. Work through it in stages: review the relevant papers, propose an approach, implement the code, compare it with [baseline], and write a brief summary of what happened, including any weak spots or uncertain parts. Stop and ask for my approval before any step that would change data, submit results, send messages, or make payments. If you cannot verify something, label it clearly so I can examine it myself afterward.

Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.

Read the original on bsky.app ↗

Get 3 agent uses like this every morning

With the day's top AI news. Free, and one click to unsubscribe.