Biology subjects

Lim, S. B.

Publications and source records attributed to Lim, S. B..

6 recordsLinked to original sources

bioRxiv2026

megaMine: a scalable, rule-based framework for mining gene-cancer-drug evidence from biomedical literature

The rapid expansion of the oncology literature has outpaced manual curation of clinically relevant gene-cancer-drug associations and oncogenic driver evidence. Existing automated approaches often lack transparency or are difficult to scale across heterogeneous data sources. To address this gap, we developed megaMine, a transparent, rule-based, and context-aware literature-mining framework that integrates therapeutic and driver evidence from PubMed, PubTator, and Europe PMC by combining entity recognition, hierarchical heuristics, and contextual labeling. In therapy mode, megaMine was applied to approximately 100,000 oncology articles published between 2015 and 2025, yielding more than 23,000 structured sentence-level evidence records, with standardized annotations for drug response, resistance, and study context. Internal evaluation of context labels showed strong separability between efficacy and non-efficacy evidence using ridge logistic regression (AUROC = 0.915; AUPRC = 0.941). Benchmarking against NCI/OncoKB-supported drug-cancer associations showed that curated clinical associations had higher megaMine composite evidence scores than unlabeled comparison pairs [median (IQR): 25.6 (9.07-72.5) vs. 3.61 (1.69-8.69); Wilcoxon rank-sum test, P < 2.2 x 10^-16]. In driver mode, megaMine retrieved mutation- and biomarker-related evidence from an ERBB-focused gastric cancer query, generating 750 evidence rows from 200 PMIDs. These results demonstrate that deterministic and interpretable approaches can support scalable evidence extraction for downstream applications such as knowledge graph construction and literature-based evidence synthesis.

bioinformatics↗

bioRxiv2023

An integrated single-cell transcriptome landscape of postnatal mouse hypothalamus

The neural stem cells (NSCs) in the hypothalamus are relatively more narrowly defined than in other neurogenic regions of the postnatal brain. By leveraging single-cell RNA sequencing (scRNA-seq) data, we generated an integrated reference dataset comprising 296,282 cells from postnatal hypothalamic regions and the adjacent region, bed nucleus of the stria terminalis (BNST), and identified 30 hypothalamic neuronal and non-neuronal cell populations. The analyses of their gene expression pattern, and specific differentiation trajectories reveal the presence of NSCs and intermediate progenitor cells (IPCs) that show a continuum of activation and differentiation processes in the hypothalamus after birth. Through comparative analyses of the integrated dataset with our lab-generated Connect-seq data obtained from the whole hypothalamus, we further assessed the technical validity of the dataset presented in this study. Our large-scale unified scRNA-seq dataset with harmonized cell-level metadata can serve as a valuable resource for investigating cell type-specific gene expression and cellular differentiation trajectories in the postnatal mouse hypothalamus.

neuroscience↗

bioRxiv2022

Deleterious Mechanical Deformation Selects Mechanoresilient Cancer Cells with Enhanced Proliferation and Chemoresistance

Cancer cells derived from secondary tumors can form new distant metastases more efficiently as compared to their primary tumor counterparts. This is partially due to the unfavorable microenvironments encountered by metastasizing cancer cells that result in the survival of a more metastatic phenotype from the original population. However, it is unclear how cancer cells may acquire such metastatic competency after overcoming deleterious mechanical stresses. Here, by forcing cancer cells to flow through small capillary-sized constrictions, we demonstrate that mechanical deformation can select a tumor cell subpopulation that exhibits resilience to mechanical squeezing-induced cell death. Transcriptomic profiling reveals up-regulated proliferation and DNA damage response pathways in this subpopulation, which are further translated into a more proliferative and chemotherapy-resistant phenotype. Our results highlight a potential link between the microenvironmental physical barriers and the enhanced malignancy of metastasizing cancer cells which may potentially be utilized for novel therapeutic strategies in preventing the metastatic spread of cancer cells.

cancer biology↗

bioRxiv2020

Cancer exosomes harbor diverse hypoxia-targeted mRNAs and contribute toward tumor angiogenesis

Exosomes are extracellular vesicles of endosome origin secreted by various cells. The exosomal cargo, especially proteins and microRNAs, have been extensively investigated for their roles in intercellular communication and as biomarkers for clinical applications. However, the understanding of types and functions of exosomal mRNAs remains limited. Here, we evaluated the mRNAs of 61 hypoxia-targeted genes in exosomes by quantitative reverse transcription PCR (RT-qPCR). Among these 61 mRNAs, 14.8% of them were detected in the MCF10A-derived exosomes, 42.6% in the MCF7-derived exosomes, and 49.2% in the MDA-MB-231-derived exosomes, many of which are differentially regulated in response to hypoxic stress in a cell-line dependent manner. Consequently, 30 exosomal mRNAs are identified as cancer related biomarkers as they are present in cancer cell-derived exosomes and absent in MCF10A-derived exosomes. Co-culture of MDA-MB-231 cells with HUVECs shows uptake of MDA-MB-231 secreted exosomes by the Human umbilical vein endothelial cells (HUVECs). Subsequently, the cancer exosomal VEGFA mRNAs were translated within the HUVECs into proteins that promoted VEGFR-dependent angiogenesis. This finding provides novel insights into how cancer cells can directly contribute towards angiogenesis. RNA-seq also shows that cancer exosomes can upregulate epithelial-mesenchymal transition-related and metabolism-related genes. The transcripts of these genes are found present in the cancer exosomes, suggesting that the uptake of exosomal mRNAs at least partially contributed to the upregulation of the corresponding mRNAs. This study shows that cancer exosomes harbor diverse mRNAs, some of which can act as promising biomarkers as well as contribute towards reprogramming of recipient cells.

cancer biology↗

bioRxiv2018

Compendiums of Cancer Transcriptome for Machine Learning Applications

BackgroundThere exist massive transcriptome profiles in the form of microarray, enabling reuse. The challenge is that they are processed with diverse platforms and preprocessing tools, requiring considerable time and informatics expertise for cross-dataset or cross-cancer analyses. If there exists a single, integrated data source consisting of thousands of samples, similar to TCGA, data-reuse will be facilitated for discovery, analysis, and validation of biomarker-based clinical strategy.\n\nFindingsWe present 11 merged microarray-acquired datasets (MMDs) of major cancer types, curating 8,386 patient-derived tumor and tumor-free samples from 95 GEO datasets. Highly concordant MMD-derived patterns of genome-wide differential gene expression were observed with matching TCGA cohorts. Using machine learning algorithms, we show that clinical models trained from all MMDs, except breast MMD, can be directly applied to RNA-seq-acquired TCGA data with an average accuracy of 0.96 in classifying cancer. Machine learning optimized MMD further aids to reveal immune landscape of human cancers critically needed in disease management and clinical interventions.\n\nConclusionsTo facilitate large-scale meta-analysis, we generated a newly curated, unified, large-scale MMD across 11 cancer types. Besides TCGA, this single data source may serve as an excellent training or test set to apply, develop, and refine machine learning algorithms that can be tapped to better define genomic landscape of human cancers.

bioinformatics↗

bioRxiv2018

Cross-platform meta-analysis reveals common matrisome variation associated with tumor genotypes and immunophenotypes in human cancers

BackgroundRecent sequencing efforts unveil genomic landscapes of the tumor microenvironment. Yet, little is known about the extent to which matrisome pattern is conserved in progressive tumors across diverse cancer types, and thus its clinical impact remains largely unexplored.\n\nFindingsUsing a newly generated, unified data resource, we conducted cross-platform assessment of a measure of altered extra-cellular matrix (ECM) composition and remodeling associated with tumor progression, termed as the matrisome index (TMI). Parallel analyses with TCGA in over 30,000 patient-derived biopsies revealed that TMI is closely associated with mutational load, tumor histopathology, and predictive of patient outcomes. We found an enrichment of specific tumor-infiltrating immune cell populations, signatures predictive of immunotherapy resistance, and several immune checkpoints in tumors with high TMI, suggesting potential role of ECM interaction with immunophenotyes and tumor immune escape mechanisms. Both epithelial cancer cells and carcinoma-associated fibroblasts are potential cellular contributors of such deregulated matrisome.\n\nConclusionsDespite wide spectrum of genetic heterogeneity and dynamic nature, matrisome abnormalities are integral to disease progression. Our resource of a curated compendium of 8,386 genome-wide profiles, molecular and clinical associations, and matrisome-tumor genotype-immunophenotype relationships identify potentially actionable immune targets that may guide personalized immunotherapy.

genomics↗