Xuhua Xia

The paper analyzes coronavirus genomes to show that those infecting tissues with antiviral proteins have CpG-deficient and U-rich genomes, suggesting adaptation to immune pressures.

The paper identifies starless bias and parameter-estimation bias in likelihood-based phylogenetic methods through analysis of site patterns in a 4-OTU case and provides R scripts for teaching.

The paper develops a method to impute missing distances in molecular phylogenetics using an approach similar to expectation-maximization, showing it outperforms the minimum evolution criterion and does not assume a molecular clock.

The paper introduces an improved method for estimating the gamma distribution shape parameter by using the F84 substitution model and simultaneous site-specific substitution estimation, implemented in DAMBE.

The paper introduces ARSDA, a software tool that reduces HTS data size and improves analysis efficiency by eliminating redundant reads and optimizing read allocation to paralogous genes.