Biology subjects

Chia, M.

Publications and source records attributed to Chia, M..

8 recordsLinked to original sources

bioRxiv2026

Accessible spatial host-microbe profiling in tumour tissues

The ability of current spatial transcriptomics platforms to sensitively and specifically detect tissue-resident microbes alongside the whole host transcriptome remains limited. Here we present HOst MicrobE Spatial-seq (HOMES-seq), a Visium-based workflow for joint spatial profiling of microbial species and the host transcriptome in formalin-fixed paraffin-embedded (FFPE) colorectal tumours. HOMES-seq incorporates an analytical framework to distinguish contaminant-derived signals from bona fide tissue-resident microbes while improving detection sensitivity and reducing sequencing costs relative to existing approaches.

microbiology↗

bioRxiv2025

Large-scale skin metagenomics reveals extensive prevalence, coordination, and functional adaptation of skin microbiome dermotypes across body sites

While skin microbiome studies have increasingly highlighted its importance in health and disease, our understanding of inter-individual heterogeneity in structure and function remains limited, impacting the ability to develop microbiome-based stratification and therapeutics. Powered by comprehensive skin microbiome characterization in a multi-ethnic population-based cohort (>3,550 shotgun metagenomes across 18 sampling sites), we established significant undescribed inter-individual heterogeneity and the extensive prevalence of distinct microbial configurations (17 species-resolution dermotypes) in seven out of nine body sites. Combining functional in silico and in vitro studies revealed insights into how these dermotypes assemble as a function of niche-dependent microbial interactions (e.g. hypoxia-dependent inhibition of S. hominis by S. epidermidis/M. luteus) and metabolic resource utilization (e.g. differential galactose and histidine metabolism). Integration of demographic, skin physiological, and behavioral data further identified >30 significant associations with host attributes. Cross-site analysis revealed remarkable coordination across disparate skin regions (predictive AUC-ROC>0.8) and bilateral consistency (Pearson {pi}>0.95), emphasizing the role of specific microbial and host factors in shaping dermotypes. Finally, we provide multiple lines of evidence that dermotype states impact the risk for skin discomfort (e.g. irritation, itch) and diseases (e.g. eczema), that when combined with our highly accurate dermotype classifiers (AUC-ROC>0.98), provide a new paradigm for understanding skin microbiome function and stratifying patients in the context of skin and other diseases.

genomics↗

bioRxiv2024

Skin metatranscriptomics reveals landscape of variation in microbial activity and gene expression across the human body

The skin microbiome plays an important role in immune homeostasis and skin health, and yet our understanding of in vivo microbial gene activity is hindered by the lack of a robust, non-invasive protocol for metatranscriptomics across skin sites. Circumventing the challenges of low microbial biomass, host contamination, and RNA stability, we developed a clinically tractable skin metatranscriptomics workflow that provides high technical reproducibility of profiles (Pearson r>0.95), uniform coverage across gene bodies, and strong enrichment of microbial mRNAs (2.5-40x). Applying this protocol to a cohort of healthy adults (n=27) across five different skin sites (n=102, paired metatranscriptomes and metagenomes), identified a striking divergence between transcriptomic and genomic abundances, with Staphylococcus species and the skin fungi Malassezia having an outsized contribution to the metatranscriptomic landscape at most sites despite their modest representation in metagenomes. Species-level analysis showed skin site-specific enrichment of gene expression (e.g. increased levels of secreted fungal phospholipase C on cheeks relative to scalp), and revealed how key pathways were transcriptionally active in vivo (e.g. propionate and 4-aminobutyrate metabolism, potentially impacting skin barrier function). Gene-level analysis identified diverse antimicrobial genes transcribed by skin commensals in situ, including several uncharacterized bacteriocins, some of which are expressed at levels comparable to known antimicrobial genes. Correlation of microbial gene expression with organismal abundances uncovered >20 genes that putatively mediate interactions between microbes (e.g. a secreted Malassezia restricta protein with strongly negative in vivo association with Cutibacterium acnes; Spearman {rho}>0.7). This work showcases the potential for leveraging skin metatranscriptomics to identify microbes whose activities play an outsized role in the community, and for uncovering pivotal microbial pathways and biomarkers linked to skin health and disease.

genomics↗

bioRxiv2022

No evidence for a common blood microbiome based on a population study of 9,770 healthy humans

Human blood is conventionally considered sterile. Recent studies have challenged this, suggesting the presence of a blood microbiome in healthy humans. We present the largest investigation to date of microbes in blood, based on shotgun sequencing libraries from 9,770 healthy subjects. Leveraging the availability of data from multiple cohorts, we stringently filtered for laboratory contaminants to identify 117 microbial species detected in the blood of sampled individuals, some of which had signatures of DNA replication. These primarily comprise of commensals associated with human body sites such as the gut (n=40), mouth (n=32), and genitourinary tract (n=18), which are species that are distinct from common pathogens detected in clinical blood cultures based on more than a decade of records from a tertiary hospital. Contrary to the expectations of a shared blood microbiome, no species were detected in 84% of individuals, while only a median of one microbial species per individual was detected in the remaining 16%. Futhermore, microbes of the same species were detected in <5% of individuals, no co-occurrence patterns similar to microbiomes in other body sites was observed, and no associations between host phenotypes (e.g. demographics and blood parameters) and microbial species could be established. Overall, these results do not support the hypothesis of a consistent core microbiome endogenous to human blood. Rather, our findings support the transient and sporadic translocation of commensal microbes, or their DNA, from other body sites into the bloodstream.

microbiology↗

bioRxiv2022

Genome-centric analysis of short and long read metagenomes reveals uncharacterized microbiome diversity in Southeast Asians

Despite extensive efforts to address it, the vastness of uncharacterized dark matter microbial genetic diversity can impact short-read sequencing based metagenomic studies. Population-specific biases in genomic reference databases can further compound this problem. Leveraging advances in long-read and Hi-C technologies, we deeply characterized 109 gut microbiomes from three ethnicities in Singapore to comprehensively reconstruct 4,497 medium and high-quality metagenome assembled genomes, 1,708 of which were missing in short-read only analysis and with >28x N50 improvement. Species-level clustering identified 70 (>10% of total) novel gut species out of 685, improved reference genomes for 363 species (53% of total), and discovered 3,413 strains that are unique to these populations. Among the top 10 most abundant gut bacteria in our study, one of the species and >80% of all strains were not represented in existing databases. Annotation of biosynthetic gene clusters (BGCs) uncovered more than 27,000 BGCs with a large fraction (36-88%) not represented in current databases, and with several unique clusters predicted to produce bacteriocins that could significantly alter microbiome community structure. These results reveal the significant uncharacterized gut microbial diversity in Southeast Asian populations and highlight the utility of hybrid metagenomic references for bioprospecting and disease-focused studies.

genomics↗

bioRxiv2019

Transcription levels of a long noncoding RNA orchestrate opposing regulatory and cell fate outcomes in yeast

Many long noncoding RNAs (lncRNAs) act in cis through transcription-coupled chromatin alterations that drive changes in local gene expression. How some cis-acting lncRNAs promote and others repress gene expression remains poorly understood. Here we report that in S. cerevisiae transcription levels of the lncRNA IRT2, located upstream in the promoter of the inducer of meiosis gene, regulate opposing chromatin and transcription states. Low IRT2 transcription displays enhancer RNA-like features. At these levels, IRT2 promotes histone exchange delivering acetylated histone H3 lysine 56 to chromatin thereby facilitating recruitment of a transcription factor and consequently activating transcription. Conversely, increasing IRT2 transcription enhances chromatin assembly and transcriptional repression. The opposing functions of IRT2 direct a regulatory circuit, which ensures that cells expressing opposite, but not one of either, mating-type loci enter meiosis. Our data demonstrate that the transcription levels of an lncRNA are key to controlling gene expression and cell fate outcomes.

biochemistry↗

bioRxiv2019

Integrated Genomic Analysis Reveals Key Features of Long Undecoded Transcript Isoform (LUTI)-based Gene Repression

Long Undecoded Transcript Isoforms (LUTIs) represent a class of non-canonical mRNAs that downregulate gene expression through the combined act of transcriptional and translational repression. While single gene studies revealed some important aspects of LUTI-based repression, how these features impact gene regulation at a global scale is unknown. By using transcript leader and direct RNA sequencing, here we identify 74 LUTI candidates that are expressed specifically during meiotic prophase. Translational repression of these candidates is ubiquitous and dependent on upstream open reading frames. However, LUTI-based transcriptional repression is highly variable. In only 50% of the cases, LUTI transcription causes downregulation of the protein-coding transcript isoform. Higher LUTI expression, enrichment of histone 3 lysine 36 trimethylation, and changes in nucleosome position are the strongest predictors of LUTI-based transcriptional repression. We conclude that LUTIs downregulate gene expression in a manner that integrates translational repression, chromatin state changes, and the magnitude of LUTI expression.

molecular biology↗

bioRxiv2018

Repression of divergent noncoding transcription by a sequence-specific transcription factor

Many active eukaryotic gene promoters exhibit divergent noncoding transcription, but the mechanisms restricting expression of these transcripts are not well understood. Here we demonstrate how a sequence-specific transcription factor represses divergent noncoding transcription at highly expressed genes in yeast. We find that depletion of the transcription factor Rap1 induces noncoding transcription in a large fraction of Rap1 regulated gene promoters. Specifically, Rap1 prevents transcription initiation at cryptic promoters near its binding sites, which is uncoupled from transcription regulation in the protein-coding direction. We further provide evidence that Rap1 acts independently of chromatin-based mechanisms to repress cryptic or divergent transcription. Finally, we show that divergent transcription in the absence of Rap1 is elicited by the RSC chromatin remodeller. We propose that a sequence-specific transcription factor limits access of basal transcription machinery to regulatory elements and adjacent sequences that act as divergent cryptic promoters, thereby providing directionality towards productive transcription.

molecular biology↗