5 min read

Testing the limits of autonomous drug design

Testing the limits of autonomous drug design
Nº 01 · The Lede arXiv Agents · Infrastructure

Red team probes AI agents that plan biosynthesis

Red team probes AI agents that plan biosynthesis
Fig. IarXiv · Filed 06 Nov 2026.

Red-teaming agentic biosynthesis planning is the subject of a new arXiv paper led by Nicholas Roehner, which attacks route-planning agents to find where their safeguards give way. Software that can plan a multi-step biosynthesis on its own has moved out of the demo stage, and a refusal trained into a single chatbot does not obviously carry over to a planner that calls design tools and databases in sequence. The approach is defensive: find the openings first, then decide where the barricades belong. It treats a design agent the way a security team treats a network, and what it offers is an attack method rather than a policy position.

Read the source →

PharmAgents2 runs drug discovery from target to candidate
Fig. IIarXiv · Filed 06 Nov 2026.
Nº 02 arXiv Drug discovery · Computational

PharmAgents2 runs drug discovery from target to candidate

PharmAgents2 runs drug discovery end to end in a new arXiv paper, using a multi-agent setup (several models with separate jobs, coordinated by a controller) to carry a program from target to candidate molecule without a person passing results between stages. It is a second-generation system aimed at the whole sequence rather than one well-defined step such as docking or property prediction. Autonomous end to end is the claim in the title. What settles it is what the assays say.

Read more →
AlphaEnzyme uses catalytic knowledge to design enzymes
Fig. IIIbioRxiv · Filed 06 Nov 2026.
Nº 03 bioRxiv Field report

AlphaEnzyme uses catalytic knowledge to design enzymes

AlphaEnzyme brings catalytic knowledge into enzyme design, using reaction mechanism and active-site chemistry to guide both the search for new enzymes and their directed evolution. The bioRxiv preprint from Y. Liu and colleagues targets a gap in protein models trained on sequence and structure alone, which capture what a protein looks like better than what chemistry it performs. Discovery and evolution are handled as one problem, with the same catalytic priors choosing which candidates to try and which mutations to make.

Read more →
Also Filed · One Brief from the queue
Nº 04 bioRxiv Structural biology · Protein design

One protein model predicts domains, sites and variants

One protein language model predicts domains, functional sites and variant effects together, in a bioRxiv preprint from P. Wang and colleagues. Three annotation tasks that labs usually run through three separate models now come out of one.

Read →

Reply with your discoveries. A human reads them. Forward freely.

Agentic Discovery  ·  Nº 137  ·  06 Nov 2026

Editor's Note

Four papers today on how much of discovery an agent can run before a human steps back in. Under two minutes.

 

Nº 01 · The Lede  —  arXiv  —  Agents · Infrastructure

Red team probes AI agents that plan biosynthesis

Red team probes AI agents that plan biosynthesis

Fig. I  arXiv · Filed 06 Nov 2026.

Red-teaming agentic biosynthesis planning is the subject of a new arXiv paper led by Nicholas Roehner, which attacks route-planning agents to find where their safeguards give way. Software that can plan a multi-step biosynthesis on its own has moved out of the demo stage, and a refusal trained into a single chatbot does not obviously carry over to a planner that calls design tools and databases in sequence. The approach is defensive: find the openings first, then decide where the barricades belong. It treats a design agent the way a security team treats a network, and what it offers is an attack method rather than a policy position.

Read the source →

Why it matters

Biosynthesis-planning agents now have a published way to be attacked, so a claim that one is safe can be checked instead of taken on trust.

 

Nº 02  —  arXiv  —  Drug discovery · Computational

PharmAgents2 runs drug discovery from target to candidate

Fig. II  arXiv · Filed 06 Nov 2026.

PharmAgents2 runs drug discovery from target to candidate

PharmAgents2 runs drug discovery end to end in a new arXiv paper, using a multi-agent setup (several models with separate jobs, coordinated by a controller) to carry a program from target to candidate molecule without a person passing results between stages. It is a second-generation system aimed at the whole sequence rather than one well-defined step such as docking or property prediction. Autonomous end to end is the claim in the title. What settles it is what the assays say.

Read more →

 

Nº 03  —  bioRxiv  —  Field report

AlphaEnzyme uses catalytic knowledge to design enzymes

Fig. III  bioRxiv · Filed 06 Nov 2026.

AlphaEnzyme uses catalytic knowledge to design enzymes

AlphaEnzyme brings catalytic knowledge into enzyme design, using reaction mechanism and active-site chemistry to guide both the search for new enzymes and their directed evolution. The bioRxiv preprint from Y. Liu and colleagues targets a gap in protein models trained on sequence and structure alone, which capture what a protein looks like better than what chemistry it performs. Discovery and evolution are handled as one problem, with the same catalytic priors choosing which candidates to try and which mutations to make.

Read more →

 

Also Filed  ·  One Brief from the queue

Nº 04  —  bioRxiv  —  Structural biology · Protein design

One protein model predicts domains, sites and variants

One protein language model predicts domains, functional sites and variant effects together, in a bioRxiv preprint from P. Wang and colleagues. Three annotation tasks that labs usually run through three separate models now come out of one.

Read →

 

· · ·

Reply with your discoveries. A human reads them. Forward freely.