Biology subjects

Chung, H. M.

Publications and source records attributed to Chung, H. M..

4 recordsLinked to original sources

bioRxiv2026

SCIMETAR-seq tracks immunophenotype, demethylation, mutations, and transcriptomes in single cells undergoing HMA therapy

5-azacytidine improves haematopoiesis and delays leukaemic progression in myelodysplastic neoplasms, but responses vary and are complicated by clonal mosaicism and heterogenous demethylation. Using a novel single-cell pipeline ("SCIMETAR-seq"), we found 5-azacytidine induced clonally distinct differentiation responses, with nuclear DNA demethylation occurring primarily in cycling progenitors. Despite the absence of significant nuclear DNA demethylation, quiescent stem cells underwent transcriptional remodelling in vivo, accompanied by 5-azacytidine-induced C*G-to-G*C mutations in mitochondria.

cancer biology↗

bioRxiv2026

Single cell multiomics reveal clonal and functional dynamics of MDS stem/progenitor cells during hypomethylating therapy

Progressive somatic mutations in hematopoietic stem cells (HSCs) drive the development of myelodysplastic neoplasms (MDS). Hypomethylating agents such as azacitidine (AZA) can improve blood counts and reduce blasts, although responses are rarely durable. Determinants of AZA response are complex and incompletely understood, although accumulating evidence suggests that epigenetic rewiring of mutated HSCs underlies improved hematopoietic output. Using single cell multiomics on longitudinal bone marrow samples, we show that AZA responsiveness involves expansion of cells with transcriptomic profiles shared with hematopoietic stem and progenitor cells (HSPCs) from healthy donors. These regenerating cells are depleted of copy number variations and of TP53 mutations. We also identify patient-restricted cell populations, some of which recede through transcriptional restoration or AZA cytotoxicity, and others which expand, regardless of initial clinical response, and dominate at progression. Individual patients carried multiple patient-restricted populations which had unique surface immunophenotypes and were genetically distinct. Strikingly, sorted cells from in vivo progression clones that were AZA-refractive in patients regained AZA-sensitivity when cultured in vitro, suggesting that lack of AZA response at the cellular level can be modulated by cell-extrinsic factors in vivo. Overall, we find that AZA response involves partial hematopoietic regeneration via functional differentiation of mutated, but not cytogenetically abnormal HSPCs, and that persistence of AZA-refractive sub-populations contributes to eventual disease progression.

cancer biology↗

bioRxiv2024

Interpretable variational encoding of genotypes identifiescomprehensive clonality and lineages in single cells geometrically

Clone assignment in single-cell genomics remains a challenge due to its diverse mutation macrostructures and many missing signals. Existing statistical methods, for the sake of numerical convergence, pose strong constraints on the form of predicted mutation patterns, so they easily identify sub-optimally fitted clones that overlook weak and rare mutations. To solve this problem, we developed SNPmanifold, a Python package that learns flexible mutation patterns using a shallow binomial variational autoencoder. The latent space of SNPmanifold can effectively represent and visualize complex mutations of SNPs (single-nucleotide polymorphisms) in the form of geometrical manifolds. Based on nuclear or mitochondrial SNPs, we demonstrated that SNPmanifold can effectively identify a large number of multiplexed donors of origin (k = 18) that all existing unsupervised methods fail and lineages of somatic clones with promising biological interpretation. Therefore, SNPmanifold can reveal insights into single-cell SNPs more comprehensively than other existing methods, especially in complex datasets.

genomics↗

bioRxiv2022

Frem1 activity regulated by Sonic Hedgehog signaling in the cranial neural crest mesenchyme guides midfacial morphogenesis

The Frem/Fras family of extracellular matrix proteins has been linked to human face shape variation and malformation, but little is known about their regulation and biological roles in facial development. During midfacial morphogenesis in mice, we observed Frem1 expression in the embryonic growth centers that form the median upper lip, nose, and palate. Expansive spatial gradients of Frem1 expression in the cranial neural crest cell (cNCC) mesenchyme of these tissues suggested transcriptional regulation by a secreted morphogen. Accordingly, Frem1 expression paralleled that of the conserved Sonic Hedgehog (Shh) target gene Gli1 in the cNCC mesenchyme. Suggesting direct transcriptional regulation by Shh signaling, we found that Frem1 expression is induced by SHH ligand stimulation or downstream pathway activation in cNCCs and observed GLI transcription factor binding at the Frem1 transcriptional start site during midfacial morphogenesis. Shh pathway antagonism reduced Frem1 expression during pathogenesis of midfacial hypoplasia, and FREM1 was sufficient to induce cNCC proliferation in a concentration-dependent manner. These findings provide novel insight into the mechanism by which the Shh pathway drives midfacial morphogenesis and reveal a functional role for Frem1 in cNCC biology that establishes the developmental basis for FREM1-associated face shape variation and malformation.

developmental biology↗