Paper List
-
Exactly Solvable Population Model with Square-Root Growth Noise and Cell-Size Regulation
This paper addresses the fundamental gap in understanding how microscopic growth fluctuations, specifically those with size-dependent (square-root) no...
-
Assessment of Simulation-based Inference Methods for Stochastic Compartmental Models
This paper addresses the core challenge of performing accurate Bayesian parameter inference for stochastic epidemic models when the likelihood functio...
-
Realistic Transition Paths for Large Biomolecular Systems: A Langevin Bridge Approach
This paper addresses the core challenge of generating physically realistic and computationally efficient transition paths between distinct protein con...
-
MoRSAIK: Sequence Motif Reactor Simulation, Analysis and Inference Kit in Python
This work addresses the computational bottleneck in simulating prebiotic RNA reactor dynamics by developing a Python package that tracks sequence moti...
-
The BEAT-CF Causal Model: A model for guiding the design of trials and observational analyses of cystic fibrosis exacerbations
This paper addresses the critical gap in cystic fibrosis exacerbation management by providing a formal causal framework that integrates expert knowled...
-
A Theoretical Framework for the Formation of Large Animal Groups: Topological Coordination, Subgroup Merging, and Velocity Inheritance
This paper addresses the core problem of how large, coordinated animal groups form in nature, challenging the classical view of gradual aggregation by...
-
ANNE Apnea Paper
This paper addresses the core challenge of achieving accurate, event-level sleep apnea detection and characterization using a non-intrusive, multimoda...
-
DeeDeeExperiment: Building an infrastructure for integrating and managing omics data analysis results in R/Bioconductor
This paper addresses the critical bottleneck of managing and organizing the growing volume of differential expression and functional enrichment analys...
Mechanistic Interpretability of Antibody Language Models Using SAEs
Department of Statistics, University of Oxford, UK | Reticular, San Francisco, USA | Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology (MIT), Cambridge, MA, USA
The 30-Second View
IN SHORT: This work addresses the core challenge of achieving both interpretability and controllable generation in domain-specific protein language models, specifically for antibody design.
Innovation (TL;DR)
- Methodology First application of Sparse Autoencoders (SAEs) to interrogate autoregressive antibody-specific language models (p-IgGen), moving beyond general protein language models.
- Methodology Systematic comparison reveals a key trade-off: TopK SAEs yield highly interpretable, monosemantic features (e.g., for CDR identity with validation accuracy 0.99) but lack causal steerability, while Ordered SAEs provide reliable generative control at the cost of interpretability.
- Biology Identifies and validates antibody-specific, biologically meaningful latent features, such as CDR identity and germline gene identity (e.g., IGHJ4 prediction with F1 macro score of 0.93), demonstrating the model's learning of immunologically relevant concepts.
Key conclusions
- TopK SAEs effectively compress and preserve biological information (CDR identity prediction accuracy 0.99 vs. 0.98 for raw neurons) and yield sparse, interpretable activation patterns localized to specific regions (e.g., CDRH3), overcoming neuron polysemanticity.
- High feature-concept correlation (e.g., F1 > 0.5 for IGHJ4 latents) does not guarantee causal steerability; steering on TopK-identified IGHJ4 features failed to consistently increase IGHJ4 proportions in generated sequences.
- Ordered SAEs, with their enforced hierarchical latent structure (via per-index nested grouping and decreasing truncation weights), successfully identify features that enable predictable generative steering, albeit with more complex activation patterns.
Abstract: Sparse autoencoders (SAEs) are a mechanistic interpretability technique that have been used to provide insight into learned concepts within large protein language models. Here, we employ TopK and Ordered SAEs to investigate an autoregressive antibody language model, p-IgGen, and steer its generation. We show that TopK SAEs can reveal biologically meaningful latent features, but high feature–concept correlation does not guarantee causal control over generation. In contrast, Ordered SAEs impose an hierarchical structure that reliably identifies steerable features, but at the expense of more complex and less interpretable activation patterns. These findings advance the mecahnistic interpretability of domain-specific protein language models and suggest that, while TopK SAEs suffice for mapping latent features to concepts, Ordered SAEs are preferable when precise generative steering is required.