R. Borges

Polymorphism-aware models in RevBayes: Species trees, disentangling Balancing Selection and CG-biased gene conversion

This paper introduces PoMoBalance, a polymorphism-aware model to disentangle balancing selection and GC-biased gene conversion using genomic data from Drosophila populations.

The patterns of codon usage between chordates and arthropods are different but co-evolving with mutational biases

This study develops a codon model to distinguish mutational biases and selection forces in chordates and arthropods, revealing differences in codon usage patterns and shared mechanisms balancing mutational biases.

Traditional phylogenetic models fail to account for variations in the effective population size

This study compares traditional phylogenetic models with population genetic models using genomic data from 17 species, finding that traditional models overlook effective population size effects despite incorporating substitution rates.

Fragment-based ab initio phasing of peptidic nanocrystals by MicroED

This paper introduces a fragment-based method for de novo structure determination of peptidic nanocrystals using MicroED, enabling the solution of previously intractable structures.

Bait-ER: a Bayesian method to detect targets of selection in Evolve-and-Resequence experiments

Bait-ER detects selection targets in Evolve-and-Resequence experiments using a Bayesian Moran model, showing high accuracy and computational efficiency.

Nucleotide Usage Biases Distort Inferences of the Species Tree

This study shows that nucleotide usage biases, like GC-bias, systematically underestimate genetic distances in species trees, affecting molecular dating of fruit fly populations.

Consistency and identifiability of the polymorphism-aware phylogenetic models

This paper proves the identifiability and consistency of polymorphism-aware phylogenetic models (PoMo) and demonstrates their effectiveness using simulated data.

Benchmarking software tools for detecting and quantifying selection in Evolve and Resequencing studies

This study benchmarks 10 software tools for detecting selection in Evolve and Resequencing studies, finding that LRT-1 performs best across scenarios.