Research map / Probabilistic and scientific modeling / Molecules and proteins
From Likelihood to Fitness: Improving Variant Effect Prediction in Protein and Genome Language Models
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Examines whether likelihood from protein and genome language models serves as a good proxy for fitness in zero-shot variant effect prediction, and proposes improvements to close the gap.
Problem
Zero-shot variant effect prediction with generative sequence models assumes that sequence likelihood is a proxy for fitness, an assumption that often breaks down in practice.
Approach
Argues that the likelihood-as-fitness assumption fails in practice and proposes methods to improve variant effect prediction in protein and genome language models beyond raw likelihood comparison. Full method details are not available from the abstract provided.
Surprising findings
- Likelihood comparison often fails as a proxy for fitness in zero-shot variant effect prediction, according to the abstract.
Closest papers by meaning
- Ultrafast classical phylogenetic method beats large protein language models on variant effect prediction (NeurIPS 2024 · similarity 0.89)
- DePLM: Denoising Protein Language Models for Property Optimization (NeurIPS 2024 · similarity 0.89)
- Steering Generative Models with Experimental Data for Protein Fitness Optimization (NeurIPS 2025 · similarity 0.88)
- Augmenting Biological Fitness Prediction Benchmarks with Landscapes Features from GraphFLA (NeurIPS 2025 · similarity 0.88)
- MutaPLM: Protein Language Modeling for Mutation Explanation and Engineering (NeurIPS 2024 · similarity 0.87)
- Training Compute-Optimal Protein Language Models (NeurIPS 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.