Biomedical subjects

Chi-Ping Day

Publications and source records attributed to Chi-Ping Day.

10 recordsLinked to original sources

PubMed2026

Knowledge-guided contextual gene set analysis with large language models.

MOTIVATION: Gene set analysis (GSA) is a foundational approach for interpreting genomic data of diseases by linking genes to biological processes. However, conventional GSA methods overlook clinical context of the analyses, often generating long lists of enriched pathways with redundant, nonspecific, or irrelevant results. Interpreting these requires extensive, ad-hoc manual effort, reducing both reliability and reproducibility. RESULTS: We introduce cGSA, a novel AI-driven framework that enhances GSA by incorporating context-aware pathway prioritization. cGSA integrates gene cluster detection, enrichment analysis, and large language models to identify pathways that are not only statistically significant but also biologically meaningful. Benchmarking on 102 curated gene sets across 19 diseases and ten disease-related biological mechanisms shows that cGSA outperforms baseline methods by over 30%, with expert validation confirming its increased precision and interpretability. Two independent case studies in melanoma and breast cancer further demonstrate its potential to uncover context-specific insights and support targeted hypothesis. AVAILABILITY AND IMPLEMENTATION: The demo website is publicly available at https://www.ncbi.nlm.nih.gov/CBBresearch/Lu/Demo/cGSA/, while the data and code can be accessed at https://github.com/ncbi-nlp/cGSA.

Large Language Models↗

PubMed2025

Long-read sequencing of single cell-derived melanoma subclones reveals divergent and parallel genomic and epigenomic evolutionary trajectories.

Tumor evolution is driven by various mutational processes, ranging from single-nucleotide variants (SNVs) to large structural variants (SVs) to dynamic shifts in DNA methylation. Current short-read sequencing methods struggle to accurately capture the full spectrum of these genomic and epigenomic alterations due to inherent technical limitations. To overcome that, here we introduce an approach for long-read sequencing of single-cell derived subclones, and use it to profile 23 subclones of a mouse melanoma cell line, characterized with distinct growth phenotypes and treatment responses. We develop a computational framework for harmonization and joint analysis of different variant types in the evolutionary context. Uniquely, our framework enables detection of recurrent amplifications of putative driver genes, generated by independent SVs across different lineages, suggesting parallel evolution. In addition, our approach revealed gradual and lineage-specific methylation changes associated with aggressive clonal phenotypes. We also show our set of phylogeny-constrained variant calls along with openly released sequencing data can be a valuable resource for the development of new computational methods.

Journal Article↗

PubMed2025

Knowledge-guided Contextual Gene Set Analysis Using Large Language Models.

Gene set analysis (GSA) is a foundational approach for interpreting genomic data of diseases by linking genes to biological processes. However, conventional GSA methods overlook clinical context of the analyses, often generating long lists of enriched pathways with redundant, nonspecific, or irrelevant results. Interpreting these requires extensive, ad-hoc manual effort, reducing both reliability and reproducibility. To address this limitation, we introduce cGSA, a novel AI-driven framework that enhances GSA by incorporating context-aware pathway prioritization. cGSA integrates gene cluster detection, enrichment analysis, and large language models to identify pathways that are not only statistically significant but also biologically meaningful. Benchmarking on 102 manually curated gene sets across 19 diseases and ten disease-related biological mechanisms shows that cGSA outperforms baseline methods by over 30%, with expert validation confirming its increased precision and interpretability. Two independent case studies in melanoma and breast cancer further demonstrate its potential to uncover context-specific insights and support targeted hypothesis generation.

Journal Article↗

PubMed2025

Severus detects somatic structural variation and complex rearrangements in cancer genomes using long-read sequencing.

For the detection of somatic structural variation (SV) in cancer genomes, long-read sequencing is advantageous over short-read sequencing with respect to mappability and variant phasing. However, most current long-read SV detection methods are not developed for the analysis of tumor genomes characterized by complex rearrangements and heterogeneity. Here, we present Severus, a breakpoint graph-based algorithm for somatic SV calling from long-read cancer sequencing. Severus works with matching normal samples, supports unbalanced cancer karyotypes, can characterize complex multibreak SV patterns and produces haplotype-specific calls. On a comprehensive multitechnology cell line panel, Severus consistently outperforms other long-read and short-read methods in terms of SV detection F1 score (harmonic mean of the precision and recall). We also illustrate that compared to long-read methods, short-read sequencing systematically misses certain classes of somatic SVs, such as insertions or clustered rearrangements. We apply Severus to several clinical cases of pediatric leukemia/lymphoma, revealing clinically relevant cryptic rearrangements missed by standard genomic panels.

Humans↗

PubMed2024

Stochastic modeling of single-cell gene expression adaptation reveals non-genomic contribution to evolution of tumor subclones.

Cancer progression is an evolutionary process driven by the selection of cells adapted to gain growth advantage. We present a formal study on the adaptation of gene expression in subclonal evolution. We model evolutionary changes in gene expression as stochastic Ornstein-Uhlenbeck processes, jointly leveraging the evolutionary history of subclones and single-cell expression data. Applying our model to sublines derived from single cells of a mouse melanoma revealed that sublines with distinct phenotypes are underlined by different patterns of gene expression adaptation, indicating non-genetic mechanisms of cancer evolution. Sublines previously observed to be resistant to anti-CTLA4 treatment showed adaptive expression of genes related to invasion and non-canonical Wnt signaling, whereas sublines that responded to treatment showed adaptive expression of genes related to proliferation and canonical Wnt signaling. Our results suggest that clonal phenotypes emerge as the result of specific adaptivity patterns of gene expression. A record of this paper's transparent peer review process is included in the supplemental information.

Animals↗

PubMed2006

Co-regulation of B-Myb expression by E2F1 and EGF receptor.

Epidermal growth factor receptor (EGFR) is a receptor tyrosine kinase that is frequently over-expressed in human cancers and is associated with tumorigenesis, and increased tumor proliferation and progression. Also found in breast tumors with high levels is B-Myb, a transcription factor whose expression is activated by E2F1/3 at the late G1 phase and the level is sustained through the S phase. Recent reports suggest a casual correlation between EGFR and B-Myb expression in primary breast carcinomas. However, the mechanism for such co-expression remains un-investigated. Here, we report that EGFR is important for B-Myb expression and the underlying mechanism involves cooperated effects from EGFR and E2F1. EGF stimulation and forced expression of EGFR significantly increase B-Myb gene activity and such increase occurs in the G1 phase. EGF-induced B-Myb expression was not significantly suppressed following inhibition of PI-3K and ERK, two major EGFR downstream pathways. In contrast, we observed EGF-induced in vivo association of nuclear EGFR to the B-Myb promoter and the association is only detected at the G1/S phase and is abolished by EGFR kinase inhibitor. As EGFR lacks DNA-binding domain but contains transactivational activity and E2F1 activates B-Myb expression in the G1/S phase, we further reasoned that nuclear EGFR might cooperate with E2F1 leading to activation of B-Myb. Indeed, we found that EGFR co-immunoprecipitated with E2F1 in an EGF-dependent manner and that EGF activated in vivo binding of E2F1 to the B-Myb promoter. Consistently, forced expression of both EGFR and E2F1 in EGFR-null CHO cells greatly enhanced B-Myb promoter activity, compared to the vector control and expression of EGFR or E2F1 alone. Promoter mutagenesis studies showed that EGF-induced activation of B-Myb promoter required both E2F and EGFR target sites. In summary, our data suggest that deregulated EGFR signaling pathway facilitate tumor cell proliferation partly via EGFR interaction with E2F1 and subsequent activation of B-Myb gene expression.

Animals↗

PubMed2005

Cancer-specific gene therapy.

Cancer cells transcriptionally activate many genes that are important for uncontrolled proliferation and cell death. Deregulated transcriptional machinery in tumor cells usually consists of increased expression/activity of transcription factors. Ideally, cancer-specific killing can be achieved by delivering a therapeutic gene under the control of the DNA elements that can be activated by transcription factors that are overexpressed and/or constitutively activated in cancer cells. Additionally, tumor-specific translation of tumor-killing genes has been also exploited in cancer gene therapy. Based on these rationales, cancer-specific expression of a therapeutic gene has emerged as a potentially successful approach for cancer gene therapy. To achieve tumor-specific expression, cancer-specific vectors are generally composed of promoters, enhancers, and/or 5'-UTR that are responsive to tumor-specific transcription factors. A number of cancer-specific promoters have been reported, such as those of probasin, human telomerase reverse transcriptase, survivin, ceruloplasmin, HER-2, osteocalcin, and carcinoembryonic antigen. Evidences suggest that the enhancer element targeted by beta-catenin can be useful to target colon cancer cells. The 5'-UTR of the basic fibroblast growth factor-2 has been reported to provide tumor specificity. Moreover, a variety of therapeutic genes demonstrated direct antitumor effects such as those encoding proapoptotic proteins p53, E1A, p202, PEA3, BAX, Bik, and prodrug metabolizing enzymes, namely thymidine kinase and cytosine deaminase. As cancerous cells of different origins vary significantly in their genetic, transcriptional/translational, and cellular profiles, the success of a cancer gene therapy will not be promised unless it is carefully designed based on the biology of a specific tumor type. Thus, tremendous research efforts have been focused on the development of non-viral vectors that selectively target various tumors resulting in minimal toxicity in the normal tissues. Significant progresses were also made in the exploitation of various novel apoptotic, cytotoxic genes as therapeutic tools that suppress the growth of different tumors. Together, these recent advances provide rationales for future clinical testing of transcriptionally targeted non-viral vectors in cancer patients.

Adenovirus E1A Proteins↗

PubMed2004

Adenoviral E1A targets Mdm4 to stabilize tumor suppressor p53.

The adenoviral protein E1A associates with multiple anticancer activities, including stabilization of p53 tumor suppressor, and has been tested through gene therapy approaches in clinical trials. To identify potential E1A-binding proteins involved in E1A's anticancer activities, we screened a yeast two-hybrid library and identified Mdm4, an Mdm2-related p53-binding protein, as a novel E1A-binding protein. The NH(2)-terminal region of Mdm4 and the CR1 domain of E1A were required for the interaction between E1A and Mdm4. E1A preferentially bound to Mdm4 rather than Mdm2 and formed a complex with p53 in the presence of Mdm4, resulting in the stabilization of p53 in a p14(ARF)-independent manner. E1A failed to stabilize p53 in the absence of Mdm4, showing that Mdm4 was required for p53 stabilization by E1A. Moreover, E1A-mediated stabilization of p53 occurred in nucleus. Although it had no effect on the p53-Mdm2 interaction, E1A facilitated Mdm4 binding to p53 and inhibited Mdm2 binding to Mdm4, resulting in decreased nuclear exportation of p53. Thus, our findings highlighted a novel mechanism, whereby E1A stabilized the p53 tumor suppressor through Mdm4.

Adenovirus E1A Proteins↗

PubMed2004

Upregulation of IKKalpha/IKKbeta by integrin-linked kinase is required for HER2/neu-induced NF-kappaB antiapoptotic pathway.

Constitutively active HER2/neu activates nuclear factor kappa-B (NF-kappaB) in cells and induces their resistance to apoptotic stimuli such as tumor necrosis factor-alpha (TNF-alpha). Here, we show that integrin-linked kinase (ILK), the crucial signal transducer in the integrin pathway, is involved in HER2/neu-mediated activation of NF-kappaB. Expression of HER2/neu increases ILK activity. Blocking ILK activity with a kinase-deficient mutant ILK (ILK-KD) inhibits NF-kappaB activation and sensitizes HER2/neu-transformed cells to TNF-alpha-induced apoptosis. Stable expression of ILK-KD in HER2/neu-transformed cells suppressed Akt phosphorylation and the expression of IkappaB kinase alpha and beta (IKKalpha and beta) at both the protein and mRNA levels, preventing IkappaB-alpha degradation and NF-kappaB activation. Furthermore, HER2/neu stimulated the transcriptional activity of the putative IKKbeta promoter through ILK and Akt. Our results demonstrate that upregulation of IKKalpha and IKKbeta by the ILK/Akt pathway is required for the HER2/neu-mediated NF-kappaB antiapoptotic pathway.

Animals↗

PubMed2002

The suppression of colon cancer cell growth in nude mice by targeting beta-catenin/TCF pathway.

The adenomatous polyposis coli (APC) or beta-catenin genes are frequently mutated in colorectal cancers, leading to activation of downstream genes with beta-catenin/T-cell factor (Tcf)-responsive promoters. We have developed a gene therapy approach selectively targeting colorectal cancer cells in which beta-catenin/Tcf4 pathway is activated by using a recombinant adenovirus AdTOP-CMV-TK, which carries a herpes simplex virus thymidine kinase gene (HSV TK) under the control of a beta-catenin/Tcf-response promoter linking to a minimum CMV promoter. AdTOP-CMV-TK and ganciclovir (GCV) treatment significantly suppressed the growth of human DLD-1 colon cancer cells in nude mice. Furthermore, no significant tumor suppression effect was observed in human hepatoma cell line SK-HEP-1, in which the beta-catenin/Tcf pathway is not activated, as a control experiment. In summary, we demonstrated the selective targeting of colorectal cancers with activated beta-catenin by AdTOP-CMV-TK and GCV treatment in animal models, as well as its therapeutic potential for colon cancer metastasized to liver.

Animals↗