What’s New on Vecura | September 28–October 2, 2026

Feedback on individual Agent responses, more control over molecular dynamics inputs, improved FPocket support, and four scientific tools to explore.

Meta description: Explore Vecura’s latest updates: Agent response feedback, molecular dynamics input selection, FPocket improvements, and four scientific tools.

Welcome to What’s New on Vecura, our regular recap of recent platform improvements and scientific capabilities.

This week, Vecura gives researchers more control over which docked complexes move into molecular dynamics simulations, introduces feedback on individual Agent responses, and improves pocket detection and chat readability. This week’s tool lineup also spans molecular conformer generation, antisense oligonucleotide research, antibody structure prediction, and retrosynthesis.

Here is what’s new from September 28 to October 2, 2026.

Agent

Help improve Vecura Agent with your feedback

Users can now give a thumbs-up or thumbs-down to individual Vecura Agent responses and select a reason when giving negative feedback.

Each rating helps the team understand which answers are useful and where the Agent needs to improve. This feedback helps guide ongoing improvements to Vecura Agent, with the aim of delivering more useful, higher-quality responses over time.

Read Agent conversations more clearly

Agent chat rendering has been improved to address an issue that could cause messages containing certain text patterns to appear truncated.

Interim thinking steps are also collapsed once a turn is complete, making it easier to focus on the final response and review the conversation.

Workflows

Choose which docked complexes continue to molecular dynamics

Researchers can now select which docked complexes feed into a molecular dynamics simulation step.

This provides more control over the transition from docking to simulation, allowing researchers to focus follow-up analysis on the complexes they choose.

Models and Tools

Improved FPocket inputs and results

The Detect with FPocket feature now accepts mmCIF input, giving researchers another supported format for pocket detection.

It also reports correctly when no pocket is found, making the outcome easier to interpret.

Three scientific tools now available

Explore four tools supporting different stages of computational research:

• OpenConf generates molecular conformers from an input 2D or 3D structure, supporting exploration of alternative molecular geometries.

• OligoAI predicts the in vitro efficacy of RNase H-mediated antisense oligonucleotides, supporting computational assessment of ASO candidates.

• RetroChimera predicts candidate reactions for synthesising a target molecule, supporting retrosynthetic analysis.

More control across computational research

This week’s updates help researchers select inputs for follow-up simulations, interpret pocket-detection results, and give feedback on Agent answers. The featured tools provide additional options for exploring molecular structures, ASO candidates, antibodies, and synthesis possibilities.

Follow What’s New on Vecura for regular updates on scientific tools, workflows, and platform capabilities.