Research
Published work, technical notes, and forthcoming whitepapers behind Reticular's genetics and embryo analysis platform.
What an embryo's genome can reveal
From the loss a chromosome count can't explain to inherited risk a child carries for life.
A deeper technical overview of Reticular's polygenic embryo analysis models, validation strategy, and family-facing report framework is in preparation.
Reticular Team
Sparse autoencoders reveal biologically meaningful antibody model features
Using sparse autoencoders to interpret an antibody language model, revealing biologically meaningful latent features and enabling precise generation control.
Rebonto Haque, Oliver M. Turnbull +4
Ordered sparse autoencoders make feature dictionaries more consistent
Introducing Ordered Sparse Autoencoders (OSAE) that establish strict latent feature ordering, improving interpretability consistency over Matryoshka baselines.
Sophie L. Wang, Alex Quach +2
Sparse autoencoders can open protein structure models to inspection
Scaling sparse autoencoders to ESM2-3B to enable mechanistic interpretability of protein structure prediction for the first time, with targeted steering of ESMFold outputs.
Nithin Parsan, David J. Yang +1