用于低样本蛋白质功能预测与设计的稀疏自编码器
Sparse Autoencoders for Low-$N$ Protein Function Prediction and Design
AI总结：
本研究系统评估了稀疏自编码器在低样本蛋白质功能预测与设计中的有效性，发现其仅用24条序列即可优于ESM2基线，并通过引导潜在变量在83%案例中生成最高适应性变体。
摘要：
从氨基酸序列预测蛋白质功能在数据稀缺（低样本）场景下仍是一个核心挑战，这限制了在仅有少量测定标记的序列-功能数据时，机器学习引导的蛋白质设计。蛋白质语言模型（pLMs）通过提供进化信息嵌入推进了该领域，而稀疏自编码器（SAEs）能够将这些嵌入分解为可解释的潜在变量，捕捉结构和功能特征。然而，SAEs在低样本功能预测和蛋白质设计中的有效性尚未被系统研究。在此，我们在多种适应性外推和蛋白质工程任务中，评估了基于微调后的ESM2嵌入训练的SAEs。我们表明，SAEs在仅使用24条序列的情况下，在适应性预测中始终优于或与ESM2基线持平，表明其稀疏潜在空间编码了紧凑且具有生物学意义的表示，能够从有限数据中更有效地泛化。此外，引导预测性潜在变量利用了pLM表示中的生物学基序，在83%的案例中产生了最高适应性的变体，相比之下，仅使用ESM2进行设计。
英文摘要：
Predicting protein function from amino acid sequence remains a central challenge in data-scarce (low-$N$) regimes, limiting machine learning-guided protein design when only small amounts of assay-labeled sequence-function data are available. Protein language models (pLMs) have advanced the field by providing evolutionary-informed embeddings and sparse autoencoders (SAEs) have enabled decomposition of these embeddings into interpretable latent variables that capture structural and functional features. However, the effectiveness of SAEs for low-$N$ function prediction and protein design has not been systematically studied. Herein, we evaluate SAEs trained on fine-tuned ESM2 embeddings across diverse fitness extrapolation and protein engineering tasks. We show that SAEs, with as few as 24 sequences, consistently outperform or compete with their ESM2 baselines in fitness prediction, indicating that their sparse latent space encodes compact and biologically meaningful representations that generalize more effectively from limited data. Moreover, steering predictive latents exploits biological motifs in pLM representations, yielding top-fitness variants in 83% of cases compared to designing with ESM2 alone.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
