Iddo Friedberg

PLUM generates antimicrobial peptides with controlled function and length, achieving high yields and diverse sequences with favorable predicted properties.

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.

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.

EpicTope predicts and validates epitope tagging sites by analyzing structural and functional features, showing effectiveness in zebrafish proteins.

The paper evaluates machine-learning models' ability to predict enzymatic functions for uncharacterized proteins and finds they fail to integrate scientific reasoning and make logic errors.

This paper develops zebrafish hand2 Cre and CreERT2 drivers using CRISPR-Cas9 to label mesodermal and ectodermal lineages, revealing lineage contributions to various tissues and vasculature.

The paper analyzes function annotation biases in the human proteome over time, revealing persistent inequality in knowledge distribution and a gap between annotation and scientific interest.

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.

This paper introduces TreasureIsland, a method for detecting genomic islands in unannotated bacterial genomes, showing competitive performance against existing predictors.

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.

The paper develops an interpretable model to predict antimicrobial resistance in Escherichia coli from livestock and companion animals using genomic and phenotypic data.

GOThresher removes annotation biases from protein function annotation datasets to improve the quality of data for machine learning models.

The paper surveys perspectives of experimental biologists, biocurators, and computational biologists on protein function prediction, highlighting differences in their views and challenges in collaboration.

The paper describes a crowdsourcing initiative, CACAO, where undergraduates contribute gene function annotations to the Gene Ontology, enhancing coverage and enabling community-driven curation.

The paper introduces MENTHU, a tool predicting MMEJ-repaired genotypes in zebrafish, showing improved sensitivity over inDelphi and validating its use for gene editing applications.

The paper develops a transfer learning model, TRAC, for antibiotic resistance class prediction and introduces a curated dataset to benchmark existing tools.

POEM identifies core operons in metagenomic data with high accuracy, demonstrating its utility in analyzing multi-gene functionalities in human gut metagenomes.

The paper introduces a model to quantify spatial dependency in gene expression using HiC and RNA-seq data, finding strong intra- and inter-chromosomal spatial correlations.

This paper characterizes ErCas12a, a new CRISPR/Cas system, showing it enables efficient DNA repair and gene integration in zebrafish and human cells.

SwiftOrtho efficiently classifies orthology across large genomic datasets with low memory usage and high accuracy.

The paper introduces a self-attention model for antibiotic resistant gene classification that improves precision and recall and detects out-of-distribution data with reliable uncertainty estimates.

The paper identifies 11 novel Drosophila genes involved in long-term memory formation using computational methods and behavioral assays.

The paper predicts novel bacteriocins from protein sequences using word embeddings that preserve amino acid order, avoiding sequence similarity-based methods.

This paper provides a high-coverage, high-confidence GO annotation for maize genes, outperforming existing annotations in accuracy using a mixed-method pipeline.

MetaRiPPquest identifies RiPPs by combining genome/metagenome mining with mass spectrometry analysis, discovering 5 novel natural products from 6 datasets.