Lightnews — Scholar-powered news

Valence Labs @valenceai.bsky.social · Jul 18

5/ If you’re attending ICML, make sure to connect with @jhartford.bsky.social and Ihab to learn more about the research happening at Valence Labs and @recursionpharma.bsky.social.

Valence Labs @valenceai.bsky.social · Jul 18

4/ “Towards Scientific Discovery with Dictionary Learning: Extracting Biological Concepts from Microscopy Foundation Models”

Link: arxiv.org/abs/2412.16247

1

Valence Labs @valenceai.bsky.social · Jul 18

3/ “A Cross Modal Knowledge Distillation & Data Augmentation Recipe for Improving Transcriptomics Representations through Morphological Features”

Link: arxiv.org/abs/2505.21317

1

Valence Labs @valenceai.bsky.social · Jul 18

2/ ”ViTally Consistent: Scaling Biological Representation Learning for Cell Microscopy”

Link: openreview.net/forum?id=niy...

1

Valence Labs @valenceai.bsky.social · Jul 18

1/ Valence Labs, @recursionpharma.bsky.social's AI research engine, is at ICML this week!

Two of our scientists, @jhartford.bsky.social and Ihab Bendidi are presenting posters on work featured in their papers 👇🧵

1 1

Valence Labs @valenceai.bsky.social · Jun 27

3/ By combining SynFlowNet with Boltz-2, this new release aims to accelerate the design of high-affinity binders for more efficient early-stage drug discovery.

Check out the GitHub repository here: github.com/recursionpha...

Learn more about Boltz-2 here: www.rxrx.ai/boltz-2

Valence Labs @valenceai.bsky.social · Jun 27

2/ Originally introduced by Valence Labs, Recursion’s AI research engine, SynFlowNet helps ensure generative models propose molecules that are not only novel but also synthetically feasible.

1

Valence Labs @valenceai.bsky.social · Jun 27

1/ We are excited to open-source the SynFlowNet-Boltz-2 trainer today! 🧵

Recursion @recursionpharma.bsky.social · Jun 27

🚀 Accelerating Boltz-2 for more efficient structure-based hit discovery. Today, Recursion is open-sourcing its SynFlowNet-Boltz trainer to enable more efficient design of high-affinity binders needed for successful early-stage drug discovery. Learn more: www.recursion.com/news/beyond-... #TechBio 🧪

Beyond Boltz-2: Toward More Powerful Drug Discovery Tools

www.recursion.com

1

Valence Labs @valenceai.bsky.social · Jun 25

Pleased to have hosted the 2025 Molecular Machine Learning Conference last week at @mila-quebec.bsky.social

Thank you to everyone who joined us and shared their valuable insights!

We hope to see you all again soon.

Recursion @recursionpharma.bsky.social · Jun 24

We had a great turnout for #MoML 2025.

Sponsored by Recursion & @valenceai.bsky.social, MoML convenes researchers from academia & industry to discuss how ML can address key scientific goals related to molecular modeling, molecular interactions & therapeutic design.

👉https://portal.ml4dd.com/ 🧪

2

Valence Labs @valenceai.bsky.social · Jun 17

Looking forward to seeing everyone at MoML this Wednesday, June 18th, at @mila-quebec.bsky.social in Montréal.

We anticipate a day of engaging discussions, talks and poster presentations, exploring research at the intersection of machine learning and drug discovery.

Agenda Here: portal.ml4dd.com

1

Valence Labs @valenceai.bsky.social · Jun 6

Check out some of the early media coverage here: www.forbes.com/sites/alexkn...

Valence Labs @valenceai.bsky.social · Jun 6

Proud to share @recursionpharma.bsky.social's exciting announcement on Boltz-2! Valence Labs, Recursion’s AI research engine, contributed to the foundational research supporting the @mit.edu team led by Regina Barzilay.

Learn more about Boltz-2 here: www.rxrx.ai/boltz-2

Recursion @recursionpharma.bsky.social · Jun 6

More on Boltz-2, the new open source AI model from MIT & Recursion capable of predicting protein binding affinity w/ unprecedented speed, scale & accuracy -- the 1st model to combine structure & binding affinity prediction, approaching FEP accuracy w/ 1000X the speed. www.youtube.com/watch?v=gRtr...

New Open Source Model from MIT & Recursion Solves Major Hurdle in AI Drug Discovery

YouTube video by Recursion

www.youtube.com

1 2

Valence Labs @valenceai.bsky.social · Jun 3

3/ MoML is an excellent opportunity to engage with leading minds, explore cutting-edge research, and connect with the community in TechBio.

We encourage everyone to secure their spot before it’s too late!

Register now to be part of MoML 2025: portal.ml4dd.com/events/molec...

Valence Labs @valenceai.bsky.social · Jun 3

2/ Our list of speakers include:
— Aaron Newman (@stanford.edu)
— Smita Krishnaswamy (@yale.edu)
— Gabriele Corso (@mit.edu)
— Kirill Neklyudov (@mila-quebec.bsky.social)
— Emmanuel Noutahi (@valenceai.bsky.social | @recursionpharma.bsky.social)
— Michael LeVine (Genesis Therapeutics)

1

Valence Labs @valenceai.bsky.social · Jun 3

1/ We're excited to announce our speaker lineup for MoML 2025!

Join us on June 18th, 2025, at Mila in Montreal for a day focused on the intersection of machine learning and drug discovery. 🧵

1 2

Reposted by Valence Labs

Recursion @recursionpharma.bsky.social · May 22

A new perspective paper from Recursion and our AI research engine @valenceai.bsky.social lays out our vision for a virtual cell as a system that can reliably drive the discovery of new drugs via an iterative loop of: predict, explain, discover. arxiv.org/abs/2505.14613 #TechBio 🧪

1 4

Valence Labs @valenceai.bsky.social · May 22

6/ Learn more

Try the app: txpert.valencelabs.com
Read the blog: www.valencelabs.com/txpert-predi...
Read the paper: arxiv.org/abs/2505.14919
Access the code: github.com/valence-labs...

TxPert: Predicting Cellular Responses to Unseen Genetic Perturbations - Valence Labs

We introduce TxPert: a state-of-the-art model that leverages multiple biological knowledge networks to accurately predict transcriptional responses under OOD scenarios.

www.valencelabs.com

1

Valence Labs @valenceai.bsky.social · May 22

5/ TxPert addresses the "Predict" capability of our virtual cell framework

This represents an initial, but important, step in a much larger journey: building models that can predict, explain, and eventually discover new therapeutic opportunities

www.valencelabs.com/advancing-dr...

Advancing Drug Discovery Outcomes with Virtual Cells at Recursion - Valence Labs

Predict, Explain, Discover: The Pillars of the Virtual Cell Our view of the virtual cell rests on three interconnected capabilities:

www.valencelabs.com

1 1

Valence Labs @valenceai.bsky.social · May 22

4/ We’re also releasing the TxPert App, a tool that makes the model accessible to researchers

- Select a cell type and gene to perturb
- Visualize the gene interaction networks
- Analyze predicted expression changes

Explore more: txpert.valencelabs.com

1 2

Valence Labs @valenceai.bsky.social · May 22

3/ TxPert achieves state-of-the-art performance by integrating multiple biological knowledge networks

These include curated public resources like STRINGdb and GO, alongside @recursionpharma.bsky.social proprietary PxMap and TxMap graphs derived from large-scale perturbation screens

1 2

Valence Labs @valenceai.bsky.social · May 22

2/ Observing how cells respond to perturbations is key to understanding disease and designing more effective therapies

But doing this experimentally across an intractable combination of genes, cell types, and conditions is slow, expensive, and infeasible

1 1

Valence Labs @valenceai.bsky.social · May 22

1/ Introducing TxPert: a new model that predicts transcriptional responses across diverse biological contexts

It’s designed to generalize across unseen single-gene perturbations, novel combinations of gene perturbations, and even new cell types 🧵

www.valencelabs.com/txpert-predi...

TxPert: Predicting Cellular Responses to Unseen Genetic Perturbations - Valence Labs

We introduce TxPert: a state-of-the-art model that leverages multiple biological knowledge networks to accurately predict transcriptional responses under OOD scenarios.

www.valencelabs.com

1 1 2

Valence Labs @valenceai.bsky.social · May 20

8/ Links below

Blog: www.valencelabs.com/advancing-dr...

Perspective paper: www.valencelabs.com/publications...

TxPert: www.valencelabs.com/publications...

Advancing Drug Discovery Outcomes with Virtual Cells at Recursion - Valence Labs

Predict, Explain, Discover: The Pillars of the Virtual Cell Our view of the virtual cell rests on three interconnected capabilities:

www.valencelabs.com

2

Valence Labs @valenceai.bsky.social · May 20

7/ This work represents an important step towards our mission of decoding biology to radically improve lives

We are excited to share our vision and invite the scientific community to explore these resources together

1 1

Valence Labs @valenceai.bsky.social · May 20

6/ By leveraging multiple biological knowledge graphs, TxPert can generalize to unseen single perturbations and combinations of novel cell types and multiple perturbations

Read the paper: www.valencelabs.com/publications...

1 1