A Diffusion Model to Shrink Proteins While Maintaining Their Function

2025-11-10

AI Summary

Protein sequences are often excessively long, limiting their applicability in pharmaceuticals and bioengineering; traditional truncation relies on labor-intensive experiments, while existing computational methods lack inductive bias for deletion operations and suffer from low search efficiency. We propose SCISOR—the first protein sequence compression model based on discrete diffusion: it simulates “noising” via forward stochastic insertion and trains a denoiser to reverse the process, implicitly learning evolutionary constraints and embedding a deletion prior. This self-supervised framework leverages large-scale evolutionary sequence data, achieving state-of-the-art performance in deletion effect prediction on ProteinGym. Compressed sequences generated by SCISOR exhibit higher fidelity to natural sequence distributions, and critical functional motifs are preserved at significantly higher rates than with existing methods. SCISOR establishes an efficient, interpretable paradigm for functional protein minimization.

Technology Category

• Search and Optimization: Evolutionary Computation
• Computer Vision: Diffusion Models for Vision
• Machine Learning: Learning on the Edge & Model Compression

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

Many proteins useful in modern medicine or bioengineering are challenging to make in the lab, fuse with other proteins in cells, or deliver to tissues in the body, because their sequences are too long. Shortening these sequences typically involves costly, time-consuming experimental campaigns. Ideally, we could instead use modern models of massive databases of sequences from nature to learn how to propose shrunken proteins that resemble sequences found in nature. Unfortunately, these models struggle to efficiently search the combinatorial space of all deletions, and are not trained with inductive biases to learn how to delete. To address this gap, we propose SCISOR, a novel discrete diffusion model that deletes letters from sequences to generate protein samples that resemble those found in nature. To do so, SCISOR trains a de-noiser to reverse a forward noising process that adds random insertions to natural sequences. As a generative model, SCISOR fits evolutionary sequence data competitively with previous large models. In evaluation, SCISOR achieves state-of-the-art predictions of the functional effects of deletions on ProteinGym. Finally, we use the SCISOR de-noiser to shrink long protein sequences, and show that its suggested deletions result in significantly more realistic proteins and more often preserve functional motifs than previous models of evolutionary sequences.

Problem

• Shrinking long proteins while maintaining biological function
• Overcoming inefficient combinatorial search in deletion optimization
• Generating realistic shortened proteins that preserve functional motifs

Innovation

• SCISOR uses discrete diffusion model for protein shrinkage
• It trains a denoiser to reverse random insertion noise
• Model generates realistic proteins preserving functional motifs