Biology subjects

Yip, K. Y.

Publications and source records attributed to Yip, K. Y..

At least 19 recordsLinked to original sources

bioRxiv2026

Spatial proximity to AD pathology defines cell-type and genotype-specific characteristics in human AD hippocampus

How amyloid-{beta} (A{beta}) plaques and tau phosphorylation spatially relate to cell response and genetic variation in Alzheimer's disease (AD) brain is currently unclear. We established a pipeline to concurrently annotate A{beta} and AT8 ptau pathology and cell-specific MERFISH expression profiles in human control and AD hippocampus. We characterized expression signatures in glia and neurons proximal or distal to A{beta} plaques, AT8 ptau-enriched neurons, or A{beta} and AT8 ptau in combination, and used spatially-defined transcriptional signatures to infer pathology-associated states in large snRNA-seq datasets lacking spatial context. We characterized variants from 120 gene loci associated with AD risk that affect cell-specific gene expression within local pathological contexts. Astrocyte APOE4/E4 expression was specifically suppressed in proximity to A{beta} plaques, whereas other APOE genotypes were unaffected by A{beta} proximity. Together, this analysis pipeline couples expression profiles in proximity to A{beta} and tau pathology, and implicates AD cis-gene variants with spatial sensitivity to AD pathology.

neuroscience↗

bioRxiv2026

Correcting spatial transcriptomics data affected by a prevalent transcript leakage problem across platforms, species, and tissues

Spatial transcriptomics has been widely applied to study the spatial distribution of cell types, cell states, and specific gene expression in tissue samples. However, we show that there is a prevalent transcript leakage problem in spatial transcriptomics data, where transcripts expressed by a cell diffuse to its neighborhood and are recurrently detected in the nearby cells. By analyzing published data sets, we show that this problem is general across data produced from different tissues and different species using different imaging-based and sequencing-based spatial transcriptomics platforms. It affects both upstream tasks such as expression quantification as well as downstream tasks such as cell-type annotation and detection of spatially-dependent gene expression. To tackle the transcript leakage problem, we propose a reference-free Bayesian model-based method, DeLeakage, which cleans up the data much more effectively than existing denoising methods. DeLeakage also improves cell-type annotation and avoids false detection of spatially dependent expression.

bioinformatics↗

bioRxiv2026

HIRA-mediated H3.3 deposition preserves hepatocyte cell identity during liver aging

Age-associated functional decline is partly driven by progressive chromatin degeneration. Maintenance of chromatin integrity preserves cell identity and promotes healthy aging, but through different mechanisms in proliferating and non-proliferating cells. However, specific mechanisms of chromatin maintenance and their compensatory capacity in proliferating and non-proliferating cells are undefined. The histone chaperone HIRA deposits the histone variant H3.3 in a DNA replication-independent manner, leading to its accumulation in aging, non-proliferating cells. Here, we show that hepatocyte-specific loss of HIRA causes loss of cell identity, metabolic dysfunction, and accelerated fibrotic pathology with age. Transcriptomic and epigenomic analyses indicate that HIRA-H3.3 preserves chromatin integrity and sustains transcription of highly expressed genes, including cell identity genes. Partial hepatectomy, associated with induced proliferation, restores identity of HIRA knockout livers with compensatory deposition of canonical histones H3.1/2. Together, these results demonstrate that HIRA-mediated H3.3 deposition is essential for safeguarding cell identity and tissue function during aging of non-proliferating cells, but this function can be rescued by tissue regeneration and associated cell proliferation. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=124 SRC="FIGDIR/small/710643v1_ufig1.gif" ALT="Figure 1"> View larger version (29K): org.highwire.dtl.DTLVardef@a2a26aorg.highwire.dtl.DTLVardef@154e5aeorg.highwire.dtl.DTLVardef@b315a7org.highwire.dtl.DTLVardef@152bca4_HPS_FORMAT_FIGEXP M_FIG C_FIG

genomics↗

bioRxiv2025

A knowledge-guided approach to recovering important rare signals from high-dimensional single-cell data

Single-cell transcriptomic data are high-dimensional, with many genes profiled in each cell. Dimensionality reduction is routinely applied to improve interpretability, remove noise and redundancy, and enable visualization. Most existing methods aim at preserving the most prominent data properties, which can lead to omission of rare but important signals. Here we propose a novel framework that uses knowledge-derived genes of interest to guide dimensionality reduction, which can help cluster rare cells and separate highly similar cell sub-populations. We demonstrate the utility of our framework in identifying endocrine cell subtypes in the pancreatic islet, highly similar hematopoietic sub-populations, and rare senescent cells.

bioinformatics↗

bioRxiv2025

Targeting CyclinD1-CDK6 to Mitigate Senescence-Driven Inflammation and Age-Associated Functional Decline

Cellular senescence contributes to aging and age-related diseases by driving chronic inflammation through the Senescence Associated Secretory Phenotype (SASP) and interferon-stimulated genes (ISGs). Cyclin D1 (CCND1), a key cell cycle regulator, is paradoxically upregulated in these non-proliferating cells. We show that CCND1 and its kinase partner CDK6 drive SASP and ISG expression in senescent cells by promoting DNA damage accumulation. This leads to the formation of cytoplasmic chromatin fragments (CCFs) that activate pro-inflammatory CGAS-STING signaling. The tumor suppressor p53 (TP53) and its target p21 (CDKN2A) antagonize this CCND1-CDK6-dependent DNA damage accumulation pathway to suppress the SASP. In aged mouse livers, senescent hepatocytes show increased Ccnd1 expression. Hepatocyte-specific Ccnd1 knockout or treatment with the Cdk4/6 inhibitor Palbociclib reduces DNA damage and ISGs in aged mouse liver. Notably, Palbociclib also suppresses frailty and improves physical performance of aged mice. These findings reveal a novel role for CCND1/CDK6 in regulating DNA damage and inflammation in senescence and aging, highlighting it as a promising therapeutic target.

cell biology↗

bioRxiv2025

Towards universal modeling of transcript isoform expression levels

A holy grail in computational biology is accurate modeling of transcript expression levels using epigenetic features, which would provide a quantitative way to study gene regulation in normal and disease states. Previous studies relied heavily on immortalized cell lines that exhibit properties different from cells in natural tissue environments. Most studies also quantified the expression of each gene by a single expression level, which fails to capture separate expression levels of different transcript isoforms of the same gene. In this study, making use of the latest large-scale dataset of paired transcriptomic and epigenomic data of human samples produced by the International Human Epigenome Consortium (IHEC), we computationally modeled the expression levels of individual transcript isoforms in 324 samples from 29 tissue types. We constructed the models using graph-based methods that integrate both location-specific epigenomic features and multiple types of gene-gene relationships. We found that to infer transcript isoform expression levels in a sample, a model that integrates information from many samples of other tissue types consistently outperforms a model trained on data from this sample itself, providing strong support that it is possible to construct a "universal" model that can accurately infer transcript isoform expression levels across tissue types.

bioinformatics↗

bioRxiv2025

SORLA upregulation suppresses global pathological effects in aged tauopathy mouse brain

A role for the trafficking receptor SORLA in reducing A{beta} levels has been well-established, however, relatively little is known with respect to whether and how SORLA can potentially affect tau pathology in vivo. Here, we show that transgenic SORLA upregulation (SORLA TG) can reverse pathological effects in aged PS19 (P301S tau) mouse brain, including tau phosphorylation and seeding, ventricle dilation, synapse loss, LTP impairment and glial hyperactivation. Proteomic analysis indicates reversion of PS19 profiles in PS19/SORLA TG hippocampus, including pathological changes in synapse-related proteins as well as key drivers of synaptic dysfunction such as Apoe and C1q. snRNA-seq analysis reveals suppression of PS19- signatures with SORLA upregulation, including proinflammatory induction of Plxnb1/Plxnb2 in glia. Tau seeding and aggregation, neuroinflammation, as well as PlxnB1/B2 induction are exacerbated in PS19 hippocampus with SORLA deletion. These results implicate a global role for SORLA in neuroprotection from tau toxicity in PS19 mouse brain.

neuroscience↗

bioRxiv2025

Few-shot in-context learning with large language models for antibody characterization

Large language models (LLMs) can learn new tasks by in-context learning (ICL), but it is unknown whether this ability reliably transfers to biological sequence classification. Here, we systematically evaluate how demonstration selection, shot count, and prompting strategies affect performance across 20 general-purpose LLMs. Using antibody characterization as a representative test case, we compare zero-shot, few-shot, and chain-of-thought (CoT) ICL on three classification tasks: humanness, antigen specificity, and isotype. Our results reveal a clear performance hierarchy: while zero-shot prompting performs near chance, few-shot prompting with randomly selected demonstrations improves performance, showing that LLMs can perform ICL using biological sequences from minimal supervision. However, matching protein-language model (pLM)-based classifier accuracy is only achieved when using label-diverse demonstrations drawn from antibodies similar to the query sequence. To leverage this insight, we introduce Sim-ICL, a framework that automatically retrieves such demonstrations. Using only 32-shot prompting, Sim-ICL achieves performance competitive with pLM-based classifiers, matching or outperforming them in two of the three tasks. Furthermore, reasoning-oriented prompts yield marginal gains and often produce fluent but biologically incorrect rationales, suggesting that current CoT explanations function as after-the-fact rationalizations rather than capturing mechanistic determinants of antibody properties. From these experiments, we derive practical design principles for ICL on biological sequences: use similarity-based, label-diverse demonstrations and modest shot counts, and treat reasoning prompts primarily as post hoc narratives rather than drivers of performance. Sim-ICL implements these principles in a streamlined, prompt-based framework for antibody sequence classification and, in principle, could be adapted to other biological sequence tasks.

bioinformatics↗

bioRxiv2024

Overcoming artificial structures in resolution-enhanced Hi-C data by signal decomposition and multi-scale attention

Computational enhancement is an important strategy for inferring high-resolution features from genome-wide chromosome conformation capture (Hi-C) data, which typically have limited resolution. Deep learning has been highly successful in this task but we show that it creates prevalent artificial structures in the enhanced data due to the need to divide the large contact matrix into small patches. In addition, previous deep learning methods largely focus on local patterns, which cannot fully capture the complexity of Hi-C data. Here we propose Smooth, High-resolution, and Accurate Reconstruction of Patterns (SHARP) for enhancing Hi-C data. It uses the novel approach of decomposing the data into three types of signals, due to one-dimensional proximity, contiguous domains, and other fine structures, and applies deep learning only to the third type of signals, such that enhancement of the first two is unaffected by the patches. For the deep learning part, SHARP uses both local and global attention mechanisms to capture multi-scale contextual information. We compare SHARP with state-of-the-art methods extensively, including application to data from new samples and another species, and show that SHARP has superior performance in terms of resolution enhancement accuracy, avoiding creation of artificial structures, identifying significant interactions, and enrichment in chromatin states.

bioinformatics↗

bioRxiv2023

Accurate identification of structural variations from cancer samples

Structural variations (SVs) are commonly found in cancer genomes. They can cause gene amplification, deletion, and fusion, among other functional consequences. With an average read length of hundreds of kilobases, nano-channel-based optical DNA mapping is powerful in detecting large SVs. However, existing SV calling methods are not tailored for cancer samples, which have special properties such as mixed cell types and sub-clones. Here we propose the COMSV method that is specifically designed for cancer samples. It shows high sensitivity and specificity in benchmark comparisons. Applying to cancer cell lines and patient samples, COMSV identifies hundreds of novel SVs per sample.

genomics↗

bioRxiv2022

The ENCODE Imputation Challenge: A critical assessment of methods for cross-cell type imputation of epigenomic profiles

Functional genomics experiments are invaluable for understanding mechanisms of gene regulation. However, comprehensively performing all such experiments, even across a fixed set of sample and assay types, is often infeasible in practice. A promising alternative to performing experiments exhaustively is to, instead, perform a core set of experiments and subsequently use machine learning methods to impute the remaining experiments. However, questions remain as to the quality of the imputations, the best approaches for performing imputations, and even what performance measures meaningfully evaluate performance of such models. In this work, we address these questions by comprehensively analyzing imputations from 23 imputation models submitted to the ENCODE Imputation Challenge. We find that measuring the quality of imputations is significantly more challenging than reported in the literature, and is confounded by three factors: major distributional shifts that arise because of differences in data collection and processing over time, the amount of available data per cell type, and redundancy among performance measures. Our systematic analyses suggest several steps that are necessary, but also simple, for fairly evaluating the performance of such models, as well as promising directions for more robust research in this area.

bioinformatics↗

bioRxiv2022

Regulatory roles of three-dimensional structures of topologically associating domains

1Transcriptional enhancers usually, but not always, regulate genes within the same topologically associating domain (TAD). We hypothesize that this incomplete insulation is due to three-dimensional structures of corresponding chromatin domains in individual cells: Whereas enhancers and genes buried inside the "core" of a domain interact mostly with other regions in the same domain, those on the "surface" can more easily interact with the outside. Here we show that a simple measure, the intra-TAD ratio, can quantify the "coreness" of a region with respect to single-cell domains it belongs. We show that domain surfaces are permissive for high gene expression, and cell type-specific active cis-regulatory elements (CREs), active histone marks, and transcription factor binding sites are enriched on domain surfaces, most strongly in chromatin subcompartments typically considered inactive. These findings suggest a "domain surface CRE" model of gene regulation. We also find that disease-associated non-coding variants are enriched on domain surfaces.

bioinformatics↗

bioRxiv2021

Quantifying full-length circular RNAs in cancer

Circular RNAs (circRNAs) are abundantly expressed in cancer. Their resistance to exonucleases enables them to have potentially stable interactions with different types of biomolecules. Alternative splicing can create different circRNA isoforms that have different sequences and unequal interaction potentials. The study of circRNA function thus requires knowledge of complete circRNA sequences. Here we describe psirc, a method that can identify full-length circRNA isoforms and quantify their expression levels from RNA sequencing data. We confirm the effectiveness and computational efficiency of psirc using both simulated and actual experimental data. Applying psirc on transcriptome profiles from nasopharyngeal carcinoma and normal nasopharynx samples, we discover and validate circRNA isoforms differentially expressed between the two groups. Compared to the assumed circular isoforms derived from linear transcript annotations, some of the alternatively spliced circular isoforms have 100 times higher expression and contain substantially fewer microRNA response elements, demonstrating the importance of quantifying full-length circRNA isoforms.

bioinformatics↗

bioRxiv2020

Simultaneous transcriptome and methylome profiles of single mouse oocytes provide novel insights on maturation and aging

BackgroundAdvanced maternal aging has become a worldwide public health issue that contributes to female fertility decline and significant risk to embryo development. Despite transcriptional and epigenetic alterations reported in oocyte maturation and development, the dynamics of gene expression and DNA dynamics associated with aging remain elusive. Here we generated simultaneous transcriptome and methylome profiles of mouse oocytes during aging and maturation at single-cell and single-base resolution to examine key biological processes and identify the key targets for novel treatment options. ResultsWe report the dynamics in transcriptome and DNA methylome in mouse oocytes during maternal aging and oocyte maturation. Age-associated gene expression changes showed mitochondrial dysfunction in GV oocytes and defects of chromosome segregation and spindle assembly in MII oocytes. EIF2 signaling protein synthesis pathway was also impaired during aged oocyte maturation. Moreover, distinctive DNA methylation patterns were demonstrated during maternal aging in GV and MII oocytes. A positive correlation between gene expression and methylation in gene body was characterized. Furthermore, we identified several promising biomarkers, including IL-7, to assess oocyte quality, which are potential therapeutic targets for improve oocyte maturation. More importantly, we built the first mouse oocyte maturation and age prediction model using transcriptome data and validated its feasibility in published data. ConclusionsThis work provides a better understanding of molecular and cellular mechanisms during mouse oocyte aging, points a new direction of oocyte quality assessment, and paves the way for developing novel treatments to improve oocyte maturation and quality in the future.

genomics↗

bioRxiv2020

Horseshoe crab genomes reveal the evolutionary fates of genes and microRNAs after three rounds (3R) of whole genome duplication

Whole genome duplication (WGD) has occurred in relatively few sexually reproducing invertebrates. Consequently, the WGD that occurred in the common ancestor of horseshoe crabs ~135 million years ago provides a rare opportunity to decipher the evolutionary consequences of a duplicated invertebrate genome. Here, we present a high-quality genome assembly for the mangrove horseshoe crab Carcinoscorpius rotundicauda (1.7Gb, N50 = 90.2Mb, with 89.8% sequences anchored to 16 pseudomolecules, 2n = 32), and a resequenced genome of the tri-spine horseshoe crab Tachypleus tridentatus (1.7Gb, N50 = 109.7Mb). Analyses of gene families, microRNAs, and synteny show that horseshoe crabs have undergone three rounds (3R) of WGD, and that these WGD events are shared with spiders. Comparison of the genomes of C. rotundicauda and T. tridentatus populations from several geographic locations further elucidates the diverse fates of both coding and noncoding genes. Together, the present study represents a cornerstone for a better understanding of the consequences of invertebrate WGD events on evolutionary fates of genes and microRNAs at individual and population levels, and highlights the genetic diversity with practical values for breeding programs and conservation of horseshoe crabs.

evolutionary biology↗

bioRxiv2020

Whole-genome analysis of noncoding genetic variations identifies multigranular regulatory element perturbations associated with Hirschsprung disease

It is widely recognized that the missing heritability of many human diseases is partially due to noncoding genetic variants, but there are multiple challenges that hinder the identification of functional disease-associated noncoding variants. The number of noncoding variants can be many times of coding variants; many of them are not functional but in linkage disequilibrium with the functional ones; different variants can have epistatic effects; different variants can affect the same genes or pathways in different individuals, and some variants are related to each other not by affecting the same gene but by affecting the binding of the same upstream regulator. To overcome these difficulties, we propose a novel analysis framework that considers convergent impacts of different genetic variants on protein binding, which provides multi-granular information about disease-associated perturbations of regulatory elements, genes, and pathways. Applying it to our whole-genome sequencing data of 918 short-segment Hirschsprung disease patients and matched controls, we identify various novel genes not detected by standard single-variant and region-based tests, functionally centering on neural crest migration and development. Our framework also identifies upstream regulators whose binding is influenced by the noncoding variants. Using human neural crest cells, we confirm cell-stage-specific regulatory roles three top novel regulatory elements on our list, respectively in the RET, RASGEF1A and PIK3C2B loci. In the PIK3C2B regulatory element, we further show that a noncoding variant found only in the affects the binding of the gliogenesis regulator NFIA, with a corresponding down-regulation of multiple genes in the same topologically associating domain.

genetics↗

bioRxiv2019

K-mer counting with low memory consumption enables fast clustering of single-cell sequencing data without read alignment

K-mer counting has many applications in sequencing data processing and analysis. However, sequencing errors can produce many false k-mers that substantially increase the memory requirement during counting. We propose a fast k-mer counting method, CQF-deNoise, which has a novel component for dynamically identifying and removing false k-mers while preserving counting accuracy. Compared with four state-of-the-art k-mer counting methods, CQF-deNoise consumed 49-76% less memory than the second best method, but still ran competitively fast. The k-mer counts from CQF-deNoise produced cell clusters from single-cell RNA-seq data highly consistent with CellRanger but required only 5% of the running time at the same memory consumption, suggesting that CQF-deNoise can be used for a preview of cell clusters for an early detection of potential data problems, before running a much more time-consuming full analysis pipeline.

bioinformatics↗

bioRxiv2018

New guidelines for DNA methylome studies regarding 5-hydroxymethylcytosine for understanding transcriptional regulation

Many DNA methylome profiling methods cannot distinguish between 5-methylcytosine (5mC) and 5-hydroxymethylcytosine (5hmC). Since 5mC typically acts as a repressive mark whereas 5hmC is an intermediate form during active demethylation, the inability to separate their signals could lead to incorrect interpretation of the data. Meanwhile, many analysis pipelines quantify methylation level by the count or ratio of methylated reads, but the proportion of discordant reads (PDR) has recently been proposed to be a better indicator of gene expression level. Is the amount of extra information contained in 5hmC signals and PDR worth the additional experimental and computational costs? Here we combine whole-genome bisulfite sequencing (WGBS) and oxidative WGBS (oxWGBS) data in normal human lung and liver tissues and their paired tumors to investigate the quantitative relationships between gene expression and signals of the two forms of DNA methylation at promoters, transcript bodies, and immediate downstream regions. We find that 5mC and 5hmC signals correlate with gene expression in the same direction in most samples, but considering both types of signals increases the accuracy of expression levels inferred from methylation data by a median of 18.2% as compared to having only standard WGBS data, showing that the two forms of methylation provide complementary information about gene expression. In addition, differential analysis between matched tumor and normal pairs is particularly affected by the superposition of 5mC and 5hmC signals in WGBS data, with at least 25-40% of the differentially methylated regions (DMRs) identified from 5mC signals not detected from WGBS data. We do not find PDR to be more informative about expression levels than ratio of methylated reads, and integrating the two types of methylation features only improves the accuracy of inferred expression levels by at most 9.8%. Our results also confirm previous finding that methylation signals at transcript bodies are more indicative of gene expression levels than promoter methylation signals, and further show that in addition to the first exon, methylation signals at the last exon and internal introns also contain non-redundant information about gene expression. Overall, our study provides concrete data for evaluating the cost effectiveness of some experimental and analysis options in the study of DNA methylation in normal and cancer samples.

genomics↗