Machine Learning in Bioinformatics

CoCoNat: a novel method based on deep-learning for coiled-coil prediction

CoCoNat predicts coiled-coil boundaries, register annotations, and oligomerization states using deep learning, outperforming existing methods on multiple metrics.

A random forest model for predicting exosomal proteins using evolutionary information and motifs

This paper develops a hybrid model combining motif-based and machine learning approaches to predict exosomal proteins with high accuracy, achieving AUROC of 0.85 on an independent dataset.

Functional profiling of the sequence stockpile: a review and assessment of in silico prediction tools

This study evaluates in silico tools for predicting molecular functions of orphan proteins, finding most methods are descriptive rather than predictive.

Turnover number predictions for kinetically uncharacterized enzymes using machine and deep learning

This paper introduces TurNuP, a model predicting enzyme turnover numbers for natural reactions, outperforming existing methods and generalizing across enzyme similarity.

General Features of Transmembrane Beta Barrels from a Large Database

The paper creates a large, high-quality database of bacterial transmembrane beta barrels using co-evolution data and presents a new algorithm with high accuracy for identifying these proteins.

DNAGPT: A Generalized Pre-trained Tool for Multiple DNA Sequence Analysis Tasks

DNAGPT pre-trains a modified generative pre-trained transformer (GPT) architecture on over 200 billion mammalian base pairs to jointly process sequence and numerical data across diverse DNA analysis tasks.

SPAT: Surface Protein Annotation Tool

SPAT annotates proteins based on their likelihood to be at the cell surface using existing annotations, showing good performance on leukemia and public datasets.

Learning Complete Protein Representation by Deep Coupling of Sequence and Structure

This paper introduces CoupleNet, a model that integrates protein sequence and structure to improve representation learning, achieving better performance in fold classification and function prediction.

xTrimoPGLM: Unified 100B-Scale Pre-trained Transformer for Deciphering the Language of Protein

A 100-billion-parameter protein language model trained with masked and autoregressive objectives together, topping prior results on 15 of 18 benchmarks and powering a fast structure predictor.

T4SEpp: a pipeline integrated with protein language models effectively predicting bacterial type IV secreted effectors

T4SEpp predicts bacterial type IV secreted effectors using protein language models and integrates multiple modules, achieving high sensitivity and specificity.

P-PPI: accurate prediction of peroxisomal protein-protein interactions (P-PPI) using deep learning-based protein sequence embeddings

This paper benchmarks deep learning-based embeddings for predicting peroxisomal protein-protein interactions using sequence data and integrates AlphaFold2-Multimer predictions for improved accuracy.

Fine-tuning Protein Embeddings for GeneralizableAnnotation Propagation

This paper fine-tunes protein embeddings to improve interpretable function prediction, showing that fine-tuned models outperform pre-trained ones in both known and novel function classification.

Assessing the performance of protein regression models

This paper evaluates how different performance metrics and generalization definitions affect the assessment of protein regression models.

Protein-Protein Interaction Prediction is Achievable with Large Language Models

This paper demonstrates that large language models can predict protein-protein interactions with high accuracy using synthetic data and fine-tuned models.

genomicBERT: A Light-weight Foundation Model for Genome Analysis using Unigram Tokenization and Specialized DNA Vocabulary

This paper introduces genomicBERT, a lightweight genome analysis model that improves efficiency and performance through specialized DNA vocabulary and tokenization, validated across diverse case studies.

NRPreTo: A Machine Learning Based Nuclear Receptor and Subfamily Prediction Tool

NRPreTo predicts Nuclear Receptor proteins and their subfamilies using enhanced features, achieving high performance and identifying 59 novel human NRs.

ULDNA: Integrating Unsupervised Multi-Source Language Models with LSTM-Attention Network for Protein-DNA Binding Site Prediction

ULDNA improves DNA-binding site prediction using a LSTM-attention network with pre-trained language models, achieving higher accuracy than existing methods on 1287 proteins.

TSignal: A transformer model for signal peptide prediction

TSignal is a transformer model that predicts signal peptides and their cleavage sites, showing improved accuracy across various organisms and peptide types.

Combining protein sequences and structures with transformers and equivariant graph neural networks to predict protein function

This paper introduces TransFun, a method that combines protein sequences and structures using transformers and equivariant graph neural networks to improve protein function prediction.

NTpred: A robust and precise machine learningframework for in-silico identification of Tyrosine nitration sites in protein sequences

NTpred improves the identification of tyrosine nitration sites in protein sequences using advanced feature extraction and ensemble classifiers, achieving better performance than existing methods on benchmark datasets.

Application of a Machine Learning Approach Towards the Targeted Identification of Phage Depolymerases

The paper develops a machine learning model to identify phage depolymerases with 90% accuracy using amino acid features.

CysPresso: A classification model utilizing deep learning protein representations to predict recombinant expression of cysteine-dense peptides

CysPresso predicts recombinant expression of cysteine-dense peptides using deep learning protein representations, finding AlphaFold2 features most effective for this task.

AnnoPRO: an innovative strategy for protein function annotation based on image-like protein representation and multimodal deep learning

AnnoPRO improves protein function annotation by using image-like representations and multimodal deep learning, showing better performance for rare protein families.

Classification of bacterial nanowire proteins using Machine Learning and Feature Engineering model

The paper develops a machine learning model to classify bacterial nanowire proteins using feature engineering, achieving high accuracy with XGBoost.

Protein Fitness Prediction is Impacted by the Interplay of Language Models, Ensemble Learning, and Sampling Methods

The paper evaluates how language models, ensemble learning, and sampling methods affect protein fitness prediction using binding affinity and thermal stability datasets.

AcrNET: Predicting Anti-CRISPR with Deep Learning

AcrNET predicts anti-CRISPR proteins using a deep learning model, achieving significant performance improvements and class-level predictions.

Serverless Prediction of Peptide Properties with Recurrent Neural Networks

The paper presents serverless deep learning models for predicting peptide properties, achieving results comparable to state-of-the-art methods without requiring cloud infrastructure.

SEQUENCE VS. STRUCTURE: DELVING DEEP INTO DATA DRIVEN PROTEIN FUNCTION PREDICTION

The paper compares sequence-based and structure-based methods for predicting protein function, finding that structure-based approaches can offer better accuracy when using high-quality protein structures.

NetGO 3.0: Protein Language Model Improves Large-scale Functional Annotations

NetGO 3.0 integrates a protein language model with NetGO 2.0 to improve functional annotations by leveraging unannotated proteins.

Fast and accurate protein function prediction from sequence through pretrained language model and homology-based label diffusion

SPROF-GO predicts protein functions from sequences using a pretrained language model and label diffusion, outperforming existing methods by over 10% in AUPR.