Ulisses Braga-Neto

PALLAS infers gene regulatory networks from time series data using penalized maximum likelihood and particle swarms, showing efficiency and accuracy in synthetic and real data.

The paper uses machine learning to identify SNVs in immune genes that distinguish sub-Saharan African and Latin American populations, suggesting a genome gradient that may explain the higher incidence of congenital Zika syndrome in Brazil.

The paper studies bias in precision estimation under separate sampling in case-control studies and proposes a corrected estimator that becomes unbiased with larger sample sizes.

The paper introduces a Bayesian classifier for 16S rRNA metagenomic data, showing it outperforms existing methods when sample size to dimensionality ratio is small.