Machine Learning in Bioinformatics

DeepAllo: Allosteric Site Prediction using Protein Language Model (pLM) with Multitask Learning

DeepAllo uses a fine-tuned protein language model with multitask learning to predict allosteric sites, achieving higher accuracy than previous methods.

Prediction of protein subcellular localization in single cells

This paper introduces PUPS, a method that predicts protein subcellular localization in single cells using protein sequences and cellular images, showing generalization to unseen proteins and cell lines.

Discovering nuclear localization signal universe through a novel deep learning model with interpretable attention units

This study introduces NLSExplorer, a deep learning model for identifying nuclear localization signals, showing improved performance on benchmark datasets and uncovering NLS patterns across 416 species.

Reliable prediction of protein-protein binding affinity changes upon mutations with Pythia-PPI

Pythia-PPI improves prediction of protein-protein binding affinity changes by using multi-task learning and self-distillation, achieving higher accuracy than existing methods.

PhosF3C: A Feature Fusion Architecture with Fine-Tuned Protein Language Model and Conformer for prediction of general phosphorylation site

The paper introduces PhosF3C, a model combining a fine-tuned protein language model and conformer architecture to predict phosphorylation sites with high accuracy.

ProBASS: a language model with sequence and structural features for predicting the effect of mutations on binding affinity

ProBASS predicts mutation effects on PPI binding affinity using sequence and structural features from ESM2 and ESM-IF1, achieving high correlation with experimental data.

In-silico identification of archaeal DNA-binding proteins

The paper develops Xenusia, a neural network tool to identify DNA-binding proteins in archaea using metagenomics data and validates its predictions experimentally.

SAKPE: A Site Attention Kinetic Parameters Prediction Method for Enzyme Engineering

SAKPE predicts enzyme kinetic parameters with improved accuracy by incorporating weighted protein representations, outperforming existing methods in kcat, Km, and Ki predictions.

AntiCP3: Prediction of Anticancer Proteins Using Evolutionary Information from Protein Language Models

AntiCP3 predicts anticancer proteins using evolutionary information and hybrid methods, achieving an AUC of 0.91.

Deep learning models for unbiased sequence-based PPI prediction plateau at an accuracy of 0.65

This study shows that sequence-based PPI prediction models using ESM-2 embeddings plateau at 0.65 accuracy, suggesting structural information may be needed for better performance.

Predicting Protein Electrostatics with Protein Language Models

This paper introduces KaML-ESM2 and KaML-ESMC models to predict protein pKa values, outperforming existing methods but facing challenges with engineered residues.

MPBind: Multitask Protein Binding Site Prediction by Protein Language Models and Equivariant Graph Neural Networks

MPBind predicts protein binding sites for multiple partners using protein language models and equivariant graph neural networks, outperforming existing methods.

ProtNote: a multimodal method for protein-function annotation

ProtNote is a multimodal deep learning model that enables supervised and zero-shot protein function prediction using free-form text inputs, generalizing to unseen functions.

PTM-Mamba: A PTM-Aware Protein Language Model with Bidirectional Gated Mamba Blocks

This paper introduces PTM-Mamba, a protein language model that incorporates PTM tokens and Mamba blocks to improve performance on PTM-specific tasks.

Protein Sequence Modelling with Bayesian Flow Networks

This paper introduces ProtBFN, a Bayesian Flow Network for protein sequence generation, which outperforms existing models and is fine-tuned for antibody generation.

Genomic Tokenizer: Toward a biology-driven tokenization in transformer models for DNA sequences

The paper introduces a biology-driven tokenizer for DNA sequences, integrates it into transformer models, and compares its performance with alternative tokenization strategies.

Accurate and Rapid Prediction of Protein pKa: Protein Language Models Reveal the Sequence-pKa Relationship

This paper introduces pKALM, a deep learning model for rapid and accurate prediction of protein pKa values, achieving high speed and performance across various residues and termini.

CaLMPhosKAN: Prediction of General Phosphorylation Sites in Proteins via Fusion of Codon Aware Embeddings with Amino Acid Aware Embeddings and Wavelet-based Kolmogorov Arnold Network

CaLMPhosKAN predicts phosphorylation sites by combining codon and amino acid embeddings with a wavelet-based KAN model, outperforming existing methods on independent test sets.

Cluster Analysis for Protein Sequences

The paper compares MMseqs2 clustering methods with traditional ML algorithms on protein sequences and finds MMseqs2 methods significantly outperform.

Predicting Kinase-Specific Phosphorylation Sites with Pretrained Protein Language Models

The paper introduces two sequence-based models to predict kinase-specific phosphorylation sites, showing improved performance and zero-shot capability.

Using Autoregressive-Transformer Model for Protein-Ligand Binding Site Prediction

This paper introduces a unified model combining a protein language model with an autoregressive transformer to predict protein-ligand binding sites, achieving broad ligand coverage and improved performance for underrepresented ligands.

KcatNet: A Geometric Deep Learning Framework for Genome-Wide Prediction of Enzyme Catalytic Efficiency

KcatNet predicts enzyme catalytic efficiency using geometric deep learning, outperforming existing methods and providing insights into enzyme-substrate interactions.

Improved Prediction of Bacterial Type VI Secretion Effector Proteins Using an Integrated Convolutional Neural Network Model Combining N-terminal Signal Sequences, Evolutionary Information and Pre-Trained Protein Language Features

This paper introduces T6CNN, a model that improves T6SE prediction by integrating signal sequences, evolutionary data, and pre-trained features, achieving higher accuracy, sensitivity, and specificity than existing tools.

Boosting Protein-Protein Interaction Detection with AlphaFold Multimer and Transformers

The paper enhances protein-protein interaction detection using a fine-tuned transformer on AlphaFold 2 Multimer outputs and benchmarks it against machine learning algorithms.

FusionESP: Improved enzyme-substrate pair prediction by fusing protein and chemical knowledge

FusionESP improves enzyme-substrate pair prediction by integrating protein and chemical language models with contrastive learning, achieving high accuracy with fewer resources.

ProToDeviseR: an automated protein topology scheme generator

ProToDeviseR automatically generates protein topology schemes from sequence data or prediction results, integrating features from multiple resources into publication-ready formats.

Extending Prot2Token: Aligning Protein Language Models for Unified and Diverse Protein Prediction Tasks

This paper extends Prot2Token to align protein language models for diverse prediction tasks, improving performance across multiple applications through pre-training and integration with ESM-2.

Seq2Topt: a sequence-based deep learning predictor of enzyme optimal temperature

Seq2Topt predicts enzyme optimal temperature using protein sequences with deep learning, achieving high accuracy and identifying key residues for temperature adaptation.

Deep generative model for protein subcellular localization prediction

This paper introduces deepGPS, a deep generative model that predicts protein subcellular localization using both sequence and image data, providing textual and image outputs.

AstraPTM: Context-Aware PTM Prediction Model for Large-Scale Proteins

AstraPTM predicts 25 PTMs in a single pass using a transformer-based framework, achieving high accuracy and sensitivity for both common and rare modifications.