Yana Bromberg

In Other Words: Learning When to Trust Generated Protein Function Descriptions

The paper introduces a method to assess the reliability of generated protein function descriptions using a judge model and shows it improves performance across multiple methods.

REFinder: Mining new enzymes from metagenomes

REFinder improves enzyme discovery from metagenomes by combining alignment-free annotation with targeted assembly, identifying more novel enzymes at lower cost than traditional methods.

Fusion2AI: functional re-annotation of the bacterial transparent matter proteins

The paper uses large language models to generate human-readable function descriptions for bacterial protein clusters, expanding functional annotations for genome and microbiome analyses.

Whole Metagenome Sequencing: not Deep Enough for Complete Microbial Function Recovery

The paper evaluates how technical factors and batch effects influence functional recovery in whole metagenome sequencing studies using oral-rinse data from Nigerian youths.

16S rRNA sequence captures microbial functional potential

The paper develops embeRNA, a method to predict microbial functions from 16S rRNA sequences without reference databases, showing better performance in novel environments.

Quantifying uncertainty in Protein Representations Across Models and Task

This study evaluates protein embeddings' reliability by introducing a model-agnostic scoring framework to quantify their ability to capture biologically relevant information.

Which pLM to choose?

This study benchmarks fourteen protein-language models (pLMs) to determine which best captures biological properties, finding that mid-scale models often outperform larger ones for general tasks.

Deciphering enzymatic potential in metagenomic reads through DNA language models

This paper introduces DNA language models to predict enzymatic potential in metagenomic reads, identifying new enzymes and expanding functional annotations.

Assembling bacterial puzzles: piecing together functions into microbial pathways

The paper introduces a co-occurrence-based method to identify microbial pathways from metagenomic data, showing improved detection of functional relationships compared to homology-based approaches.

Learning from the unknown: exploring the range of bacterial functionality

This paper introduces a novel method to explore bacterial functional repertoires without reference databases, showing it outperforms existing approaches in taxonomic assignment and captures environment-driven diversity.

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.

mebipred: identifying metal-binding potential in protein sequences.

The paper introduces mebipred, a machine learning tool for predicting metal-binding proteins from sequence data, achieving 90% accuracy and outperforming existing methods.

Shedding Light on Microbial Dark Matter with A Universal Language of Life

This paper introduces LookingGlass, a deep learning model for short-read DNA sequences that captures functional and evolutionary features, enabling tasks like identifying novel enzymes and predicting optimal temperatures.

Decoding the effects of synonymous variants

The paper introduces synVep, a machine learning method for predicting the effects of synonymous variants, achieving high precision and recall while improving differentiation between pathogenic and benign variants.

PredictProtein - Predicting Protein Structure and Function for 29 Years

PredictProtein provides protein sequence analysis tools, offering structure and function predictions with updates including deep learning and DNA/RNA/protein binding predictions.

Evolution of the SARS-CoV-2 proteome in three dimensions (3D) during the first six months of the COVID-19 pandemic

The paper analyzes 3D structural changes in SARS-CoV-2 proteins during the first six months of the pandemic, showing amino acid substitutions primarily affect surfaces and boundary layers.

Snow Microbiome Functional Analyses Reveal Novel Microbial Metabolism of Complex Organic Compounds

The paper analyzes snow microbiome metagenomes and metatranscriptomes to reveal functional diversity and activity, finding higher diversity in early spring and correlations between microbial pathways and organic acid levels.

Identifying Crohns disease signal from variome analysis

This paper introduces AVA,Dx, a machine learning method using whole exome sequencing to predict Crohn's disease status with 71% precision and 73% recall.

fuNTRp: Identifying protein positions for variation driven functional tuning

The paper introduces funtrp, a method to classify protein positions as neutral, rheostat, or toggle based on mutational impact, showing these types correlate with functional effects and improve variant effect prediction.

fusionDB: assessing microbial diversity and environmental preferences via functional similarity networks

fusionDB links microbial functions to environmental metadata through functional similarity networks, enabling comparisons and identification of horizontal gene transfer events.

Computational prediction shines light on type III secretion origins

The paper introduces a computational method to identify type III effectors in bacteria, finding that recognition signals span the entire protein sequence and identifying new effectors across prokaryotic genomes.