Layer Probing Improves Kinase Functional Prediction with Protein Language Models
Protein language models (PLMs) have transformed sequence-based protein analysis, yet most applications rely only on final-layer embeddings, which may overlook biologically meaningful information encoded in earlier layers. We systematically evaluate all 33 layers of ESM-2 for kinase functional prediction using both unsupervised clustering and supervised classification. We show that mid-to-late transformer layers (layers 20-33) outperform the final layer by 32 percent in unsupervised Adjusted Rand Index and improve homology-aware supervised accuracy to 75.7 percent. Domain-level extraction, calibrated probability estimates, and a reproducible benchmarking pipeline further strengthen reliability. Our results demonstrate that transformer depth contains functionally distinct biological signals and that principled layer selection significantly improves kinase function prediction.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Leveraging Protein Language Model Embeddings for Catalytic Turnover Prediction of Adenylate Kinase Orthologs in a Low-Data Regime
Accurate prediction of enzymatic activity from amino acid sequences could drastically accelerate enzyme engineering for applications such as bioremediation and therapeutics development. In recent years, Protein Language …
Language ModelingLanguage ModellingPredictionProtein Language ModelQuantitative and functional post-translational modification proteomics reveals that TREPH1 plays a role in plant thigmomorphogenesis
Plants can sense both intracellular and extracellular mechanical forces and can respond through morphological changes. The signaling components responsible for mechanotransduction of the touch response are largely unknow…
Giant Hub Src and Syk Tyrosine Kinase Thermodynamic Profiles Recapitulate Evolution
Thermodynamic scaling theory, previously applied mainly to small proteins, here analyzes quantitative evolution of the titled functional network giant hub enzymes. The broad domain structure identified homologically is c…
Multi-layer attentive probing improves transfer of audio representations for bioacoustics
Probing heads map the representations learned from audio by a machine learning model to downstream task labels and are a key component in evaluating representation learning. Most bioacoustic benchmarks use a fixed, low-c…
Representation LearningDeepGuard: Secure Code Generation via Multi-Layer Semantic Aggregation
Large Language Models (LLMs) for code generation can replicate insecure patterns from their training data. To mitigate this, a common strategy for security hardening is to fine-tune models using supervision derived from …
Code Generation