Twitxy Saved

IMPRINT

Verified account This account is verified. Learn more

@IMPRINTlabs

IMPRINT’s mission is to create forensic immunology research tools that are scalable, cost-efficient, and accessible to the biomedical community.
New York, New York, United Sta imprint.org Joined November 2024
20 Following 169 Followers 22 Posts

Latest posts 5

IMPRINT Verified account @IMPRINTlabs
At IMPRINT, we’re building the tools to decode the immune system’s record.

This @Spec__Tech profile highlights how our CEO & co-founder @BeckBrachman – a Brains Research Accelerator Fellow – & IMPRINT is advancing foundational immunomics to unlock new paths in diagnostics and therapeutics.

Thank you @Ben_Reinhardt & @lr_bio for the thoughtful profile

🔗 Read the full article: blog.spec.tech/p/beck-brachma…
blog.spec.tech Beck Brachman's Quest to Decode the Immune System The first post in a mini-series about the Brains Fellows
IMPRINT Verified account @IMPRINTlabs
Thank you to @ChanZuckerberg and @officialbiohub for convening the AI × Immunology Affiliate Symposium in NYC.

Our CEO & co-founder, @BeckBrachman, presented alongside leaders advancing the future of R&D, technology, and human health, contributing to discussions on emerging approaches in AI-enabled immunology.

Collaborative forums like this play an important role in driving scientific progress.
IMPRINT Verified account @IMPRINTlabs
Models trained on harder antibody binding tasks learned deeper biological rules and generalized better.

A recent @NatMachIntell paper by our cofounder & CTO @victorgreiff shows training on hard negatives (weak vs strong binders) boosts antibody ML generalization.

At IMPRINT, this mindset drives our push for robust AI-based mapping of adaptive immunity.

nature.com/articles/s4225…

#MachineLearning #Immunology #AntibodyDiscovery #AIForScience
www.nature.com Training data composition determines machine learning generalization and biological rule discovery Nature Machine Intelligence - Negative data composition critically shapes machine learning robustness in sequence-based biological tasks. Training data composition and its implications are...
IMPRINT Verified account @IMPRINTlabs
Our CEO & Co-Founder @BeckBrachman spoke at #MoForum in NYC on how AI/ML is reshaping immunology and the opportunities ahead for life sciences — alongside amazing innovators 🔬

A huge thank you to @westendlabs and @MoFoLLP for bringing together leaders across science, technology and business to push the boundaries of what's possible in NYC! 🌆

#MachineLearning #Immunology #AntibodyDiscovery #AIForScience