Feiyang Ma

This study shows that aberrant oxidative metabolism drives the selection of TET2-deficient hematopoietic stem and progenitor cells in clonal hematopoiesis, with the pentose phosphate pathway playing a compensatory role.

This paper shows that targeting STAT3 with a degrader can overcome Venetoclax resistance in AML by reducing MCL1 and improving mitochondrial function.

This study identifies TREM2+ lipid-laden macrophages in alveolar TB pneumonia that support Mtb survival, contrasting with granulomas, and shows pharmacological reversal of this immunometabolic state.

This study compares dermatomyositis and lupus skin lesions using single-cell RNA-sequencing to identify inflammatory monocytes and JAK-1 signaling as drivers of vasculopathy in dermatomyositis.

The paper identifies ZBP1 and mitochondrial Z-DNA as drivers of autoimmune photosensitivity through cGAS-STING activation in response to UVB.

This study performs a GWAS meta-analysis of psoriasis, identifying 109 susceptibility loci, including 45 new ones, and highlights potential therapeutic targets and disease mechanisms.

The paper identifies biological principles governing postnatal aortic growth, showing it scales with the vertebral column and relies on differentiated cell proliferation rather than stem cells.

The paper identifies PFKFB3-expressing dysfunctional β-cells in T2D, showing their survival via positive epistasis and reduced clearance, and demonstrates PFKFB3 inhibition restores glucose tolerance in mice.

The paper identifies heterogeneous pdgfrβ+ cells critical for coronary vessel development and heart regeneration in zebrafish.

The paper uses single-cell RNA sequencing to study ductus arteriosus closure in mice, identifying vimentin as critical for vascular remodeling and closure.

This study evaluates ten deconvolution tools on 4,000 clinical and synthetic samples, finding EPIC performs best on Framingham data and HPCA references are generally accurate.

GEDIT estimates cell type abundances from gene expression data, outperforming existing tools across multiple platforms and tissue types.

Single-cell RNA sequencing reveals that heterogeneity in Chlamydomonas bulk cultures is largely driven by diurnal cycle phases.

ACTINN is a neural network that automatically identifies cell types in scRNA-seq data by learning features from annotated datasets, outperforming manual methods in speed and accuracy.

The paper introduces MREBS, a method for DNA methylation estimation that reduces sequencing costs while improving coverage compared to existing methods.