Xionglei He

PASTRI infers stage-specific cell-state transition rates from annotated cell lineages, revealing rate-limiting steps and attractor states in development and disease.

This paper reconstructs whole-organism cell phylogenies with resolved ancestral transcriptional states in zebrafish larvae using a novel computational framework called LEAP.

The paper identifies high mutation rates in histone genes due to AID/APOBEC enzymes and introduces HistMTR to predict pathogenicity, showing histone mutations contribute to diseases.

This study uses lineage tracing and single-cell analysis to show tumor plasticity precedes stable immune evasion, with AP-1 regulating EMT stabilization during immunotherapy.

The paper introduces a dimension decomposition method to separate genetic and non-genetic factors in phenotype spaces, showing that traits are coded by a limited number of genetic dimensions and trait-specific non-genetic dimensions.

scPhyloX infers tissue development and tumor evolution dynamics using single-cell phylogenetic trees, revealing novel insights into stem cell overshoot, hematopoietic progenitor clonal expansion, and colorectal tumor subclonal selection.

This study identifies and quantifies developmental canals in C. elegans embryogenesis that suppress noise and stabilize development, revealing how cell-level stochasticity affects phenotypic outcomes.

The paper introduces PhyloVelo, a method that uses monotonically expressed genes to reconstruct cell fate transitions, outperforming existing approaches in embryo development and tumor evolution analysis.

This paper introduces a coalescent method, i-coalescent analysis, that is insensitive to population structure and demonstrates its ability to accurately reconstruct population size dynamics in structured populations.

This study uses scLOESS to trace lineage commitments in zebrafish embryos, revealing development histories and ongoing lineage events through single-cell analysis.

The paper shows that gene deletion effects can be split into clustered and distributed effects, with only clustered effects aligning with known biochemical functions and distributed effects being evolutionarily decoupled.

This study examines the structure of a yeast phenotype space, revealing convergent dimensions and stratification between genotype and phenotype spaces.

The paper analyzes sex ratio dynamics in Tetrahymena thermophila using a population genetics model and experimental validation, showing support for theoretical predictions about sex coexistence and fixation.

The paper shows that DNA curvature affects mutation rates in yeast and human tumors, with lower curvature linked to higher mutation rates.

The paper introduces a measure of trait tractability (ST) based on gene expression states in yeast, showing it is constrained by natural selection and predicts the determinability of traits from expression data.

The paper investigates how natural selection shapes the regulator-executor-phenotype architecture in yeast by analyzing morphological traits and their genetic control.

The paper identifies type II essential genes, which can become non-essential through secondary loss-of-function mutations, suggesting a therapeutic strategy for human diseases like ADSL deficiency.

The paper analyzes 500 yeast traits to show observational approaches work best for traits under strong selection, while genetic approaches succeed when coordinating genes are perturbed.

The paper identifies expression informative genes (EIGs) correlated with fitness traits in yeast, showing natural selection shapes simple functional architectures for fitness-coupled traits, while complex epistasis governs fitness-uncoupled traits.