Machine Learning in Bioinformatics

ProtBLIP2-SST: Protein Function Prediction via BLIP2 with Sequence, Structure, and Text

ProtBLIP2-SST generates open-ended protein function captions by integrating sequence, structure, and text using a two-stage framework.

ProtAug: An Empirical Investigation of pLM-Guided Data Augmentation for Protein Sequence Prediction Tasks

This paper evaluates pLM-guided data augmentation for protein sequence prediction, showing that it improves performance across tasks with variation levels between 2-30%.

MAERM: Predicting Enzyme-Reaction Matching Relationships with a Mixed-Attention Model

MAERM predicts enzyme-reaction matching relationships using a mixed-attention model, outperforming baselines and enabling enzyme design.

Emergence of Biological Structural Discovery in General-Purpose Language Models

This paper shows that general-purpose language models can detect protein homology without biological pre-training, achieving ROC-AUC 0.79 on a controlled benchmark.

PINPOINT: Protease INhibitor PredictiOn at the plant-pathogen INTerface using protein language models and structural modeling

PINPOINT predicts protease-inhibitory activities of fungal SSPs lacking annotated inhibitor domains using protein language models and structural modeling.

SAPPTree: Identification of an S-Acylation Motif Drives a Novel S-Acylation Prediction Program

The paper develops a new S-acylation prediction tool, SAPPTree, by integrating sequence and structural features from experimentally confirmed sites to improve prediction accuracy.

ProLoc: Text-guided Localization of Protein Functional Regions

ProLoc localizes protein functional regions using text guidance, outperforming existing methods on a new benchmark.

Performance Attribution in pLM-Based Biological Relation Prediction

The paper evaluates pLM-based biological relation prediction models to determine the sources of their performance, finding that multiple information sources contribute and that aggregate metrics alone cannot establish mechanistic insights.

How Not to be Seen: Predicting Unseen Enzyme Functions using Contrastive Learning

This paper introduces EnzPlacer, a contrastive learning method that predicts enzyme function from sequence by placing it in known function space, even when the precise function is unknown.

Optimizing Protein Tokenization: Reduced Amino Acid Alphabets for Efficient and Accurate Protein Language Models

This study explores using reduced amino acid alphabets with sub-word tokenization in protein language models, showing improved efficiency with minimal performance loss.

ACCURATE PREDICTION OF ASPARAGINE DEAMIDATION IN BIOLOGICS USING ADVANCED MACHINE LEARNING MODELS

This study develops machine learning models to predict asparagine deamidation in proteins, showing improved accuracy over existing methods and reducing development costs.

AbTune: Layer-wise selective fine-tuning of protein language models for antibodies

AbTune improves protein language model performance for antibody-related tasks through depth-selective fine-tuning, outperforming baselines in structure prediction and mutation effect prediction.

Probe-Based Identification of Metal-Binding Sites Using Deep Learning Representations

PRIME is a deep learning framework that predicts metal-binding sites in proteins by integrating evolutionary and structural signals, achieving high accuracy across various metal ions.

EnzyKAN: Protein Language Model Embeddings and Kolmogorov-Arnold Network Variants for Enzyme Commission Classification with a Proposed Electron-Transfer Physics Feature Framework

The paper evaluates KAN variants and protein embeddings for enzyme classification, showing improved performance with ESM-2 embeddings.

Prot2Prop: Structure-informed multitask protein property prediction

Prot2Prop predicts six protein developability properties using a multitask framework with a frozen ProstT5 encoder and task-specific adapters, achieving strong performance on both classification and regression tasks.

Choragraph: A deep-learning approach for the analysis of spatial proteomics reveals subcellular Arabidopsis protein trafficking routes and multi-residency

Choragraph uses deep learning to analyze spatial proteomics data, reconstructing missing values and identifying subcellular localization, including dual-residency in Arabidopsis.

LAMBDA: A Prophage Detection Benchmark for Genomic Language Models

The paper introduces LAMBDA, a benchmark to evaluate genomic language models' ability to detect prophage sequences across four complexity levels.

seqLens: optimizing language models for genomic predictions

This study develops seqLens, a genomic language model family, and evaluates its performance on phenotypic predictions using diverse pretraining datasets and tokenization strategies.

A Convolutional Deep Learning Approach to identify DNA Sequences for Gene Prediction

The paper presents a convolutional neural network for gene prediction using human genome sequences, achieving state-of-the-art results on genes associated with genetic disorders.

GUANinE v1.1 Reveals Complementarity of Supervised and Genomic Language Models

GUANinE v1.1 evaluates supervised and unsupervised genomic models, showing each excels on specific tasks and identifying tradeoffs in model performance and resource use.

A PLM-Based Method for Predicting Protein Ion Channel Modulators for Drug Discovery and Safety Evaluation

This paper introduces IonNTXpred, a PLM-based method for predicting ion channel modulators, achieving high accuracy and supporting drug discovery and safety evaluation.

A unified smoothing framework for protein domain bigram model

The paper introduces a unified smoothing framework for protein domain bigram models, distinguishing between biologically incompatible and undersampled domain pairs.

EssTFNet: Integration of Adaptive Time - Frequency and DNA Language Models for Interpretable Human Essential Gene Prediction

EssTFNet predicts human essential genes using a deep learning framework combining adaptive time-frequency analysis and DNA language models, achieving high accuracy and interpretability.

Handshake: Partner-Specific Protein-Protein Binding Site Prediction at Scale Using ProstT5 and Cross-Chain Attention

HandShake predicts partner-specific protein-protein binding sites using only sequence data, outperforming existing methods and revealing redundancy inflation in benchmark metrics.

Biological meaning in protein embedding space is resolution-dependent

The paper shows that biological meaning in protein embeddings depends on resolution, with different levels of organisation preserving distinct biological relationships.

SPPIDER-seq: Sequence-based partner-aware predictor of protein-protein interaction sites

SPPIDER-seq predicts protein-protein interaction sites by considering partner-specific context, outperforming AlphaFold3 on disordered interfaces and revealing partner-specific patterns in TP53 interactions.

ProtAff: Protein Binding Affinity Prediction via LoRA-Finetuned ESM-2

ProtAff predicts protein-protein binding affinity using sequence data, outperforming existing methods on an EGFR benchmark but underperforming structural methods in cross-target discrimination.

Sequence-Based Therapeutic Peptide Classification with Augmented Negative Sampling

This paper introduces a CNN classifier for therapeutic peptide classification using augmented negative sampling, achieving significantly lower false positive rates and competitive performance on external benchmarks.

Explainable protein-protein binding affinity prediction via fine-tuning protein language models

BALM-PPI predicts protein-protein binding affinity from sequence using a fine-tuned protein language model, achieving accuracy comparable to structure-based methods and providing residue-level explanations.

Advances in protein function prediction from the fifth CAFA challenge

This work analyzes CAFA5 submissions, showing improved accuracy in protein function prediction and highlights the role of protein language models and 3D structures in performance gains.