Sean R. Eddy

The paper identifies MOSN, a novel sex-specifically spliced neuronal gene in Aedes aegypti, conserved in other mosquito species but absent in Drosophila, with distinct protein isoforms and a duplicated domain.

The paper induces menstruation in mice to study endometrial shedding regulation, revealing a new paradigm for spatially regulated fibroblast differentiation.

This study identifies group II introns in phage genomes using RNA structure-based homology searches, revealing their prevalence across diverse phage lineages.

The paper builds a cellular atlas of the hypothalamic preoptic area in two deer mouse species, identifying cell types linked to behavioral differences and sex-specific evolution.

PSALM annotates protein domains using a language model, achieving sensitivity-specificity tradeoff comparable to HMMER on a large benchmark.

ORFeus detects non-canonical translation events from ribosome profiling data, identifying known alternative open reading frames and enabling novel event discovery across organisms.

The paper shows that using different gene annotation methods in comparative genomics can falsely increase the number of lineage-specific genes by up to 15-fold.

The paper introduces two new methods for splitting biological sequence data into training and test sets to improve benchmarking by ensuring test sequences are dissimilar to training sequences.

The paper develops a computational method to identify alternative genetic codes in bacterial and archaeal genomes, discovering new codon reassignments and their evolutionary drivers.

This paper introduces a hidden Potts model for sequence alignment and homology search, showing promising results in RNA structure homology benchmarks.

The paper challenges the validity of covariation analysis in lncRNA structures by showing that the RAFS statistic can falsely identify covarying base pairs due to primary sequence conservation.

The paper assesses the power of sequence covariation to detect conserved RNA structure and finds that some previously unconfirmed RNAs may have sufficient covariation signal.

The paper characterizes gene expression in 67 Drosophila visual system cell types using TAPIN-seq and genetic drivers, building a transcriptome resource to study neural circuit function.

This study investigates how CNN architecture affects the learning of genomic sequence motifs, finding that hierarchical designs lead to distributed representations while non-hierarchical designs yield interpretable localist representations.

This paper introduces ResidualBind, a convolutional residual network that infers RNA-binding protein sequence and structure specificities from enriched sequence data, outperforming previous methods and learning RNA structure context from sequence data.