Biology subjects

Hsieh, P.-H.

Publications and source records attributed to Hsieh, P.-H..

14 recordsLinked to original sources

bioRxiv2025

Pan-cancer analysis of patient-specific gene regulatory landscapes identifies recurrent PD-1 pathway dysregulation linked to outcome

Cancer heterogeneity remains a major challenge to the development of broadly effective therapies. Although gene expression-based stratification has advanced our understanding of molecular subtypes, it often overlooks the regulatory mechanisms driving tumor progression and therapeutic resistance. These mechanisms are governed by gene regulatory networks--complex systems of transcription factors and other regulators that control gene expression. To investigate regulatory heterogeneity at the patient level, we inferred individual gene regulatory networks for over 9,000 tumors across 33 cancer types from The Cancer Genome Atlas enabling a comprehensive pancancer analysis of inter-patient regulatory variation. Our analysis uncovered novel regulatory subtypes and revealed recurrent dysregulation of the PD-1 signaling pathway in 23 cancer types. This dysregulation was associated with patient outcomes in 12 cancers and appeared largely independent of established molecular subtypes and immune cell composition, suggesting alternative regulatory mechanisms. We identified both well-characterized (FOS, JUNB, ATF3) and previously unrecognized (RFX7, ELF1, NFYA, NFYC) transcriptional regulators of PD-L1 (CD274), highlighting new candidate targets for modulating immune checkpoint activity across tumor types. Our work demonstrates the power of single-sample gene regulatory networks to uncover hidden layers of tumor biology, improve patient stratification, and inform future research in immunotherapy.

bioinformatics↗

bioRxiv2025

MAESTRO uncovers tandem paralog dynamics as a core driver of fungal stress adaptation

Understanding how fungi adapt to diverse stresses is critical for mitigating emerging drug resistance and harnessing their robustness for biotechnology. Pichia kudriavzevii is a stress-tolerant and intrinsically drug-resistant yeast, with dual industrial and clinical importance. We analyzed 170 strains using MAESTRO, a machine-learning-assisted GWAS pipeline optimized for small cohorts. MAESTRO identified biologically meaningful features and revealed that copy number variation (CNV) of tandem paralogs (TPs) is a core mechanism of multi-stress adaptation. TPs were recurrently linked to tolerance of industrial inhibitors (HMF, phenolics, heat) and antifungal drugs (fluconazole, azoles), and deletion of the TP pair gene4260/gene4261 confirmed pleiotropic effects across stresses. These findings support a TP CNV model where recombination-driven TP CNVs and gene fusions enable rapid stress adaptation. Importantly, our results suggest that antifungal resistance can arise through co-option of mechanisms originally evolved for environmental stressors, raising a One Health concern about the environmental origins of drug-resistant pathogens.

genomics↗

bioRxiv2024

Gene regulatory network integration with multi-omics data enhances survival predictions in cancer

The emergence of high-throughput omics technologies has resulted in their wide application to cancer studies, greatly increasing our understanding of the disruptions occurring at different molecular levels. The role of gene regulation as a core driver of biological processes and its role in the development and progression of cancer has been well established. Transcriptional regulation, a crucial aspect of gene regulation, can be represented as a network of interactions between regulators (such as transcription factors) and their target genes. These networks are known as gene regulatory networks (GRNs). Several joint dimensionality reduction (JDR) methods and tools for integrating multi-modal data have been developed in recent years. This study presents a new approach to integrate multi-modal data of different dimensions to consider sample-specific GRNs with multi-omics data in ten cancer datasets from The Cancer Genome Atlas (TCGA). We compare JDR models that include GRNs with those that do not, assessing their association with patient survival. We find that including GRNs can improve associations with patient survival across several cancer types. Focusing on liver cancer, our methodology identifies potential mechanisms of gene regulatory dysregulation associated with cancer progression.

systems biology↗

bioRxiv2023

A Novel Genome Optimization Tool for Chromosome-Level Assembly across Diverse Sequencing Techniques

This paper introduces a novel genome assembly optimization tool named LOCLA, which stands for "Local Optimization for Chromosome-Level Assembly". It identifies reads and contigs aligned locally with high quality on gap flanks or scaffold boundaries of draft assemblies for gap filling and scaffold connection. LOCLA applies to both de novo and reference-based assemblies. It can also utilize reads produced by diverse sequencing techniques, e.g., 10x Genomics (10xG) Linked-Reads, and PacBio HiFi reads. We validated LOCLA on three human samples and one non-model organism. For the first two human samples, LLD0021C and CHM13, we generated de novo draft assemblies from 10xG Linked-Reads. On LLD0021C, LOCLA improves the draft assembly by adding 23.3 million bases using only 10xG Linked-Reads. These additional bases cover 28,746 protein-coding regions, particularly in pericentromeric and telomeric regions. On the CHM13 sample, we took 10xG Linked-Reads and PacBio HiFi reads as input. As a result, LOCLA added 46.2 million bases to the draft assembly. The increased content enables us to identify genes linked to complex diseases (e.g., ARHGAP11A) and critical biological pathways. We created two reference-guided draft assemblies on the third human sample, HG002, using contigs assembled from PacBio HiFi reads. LOCLA enhances the two draft assemblies by recovering 27.9 million bases (22.26%) and 35.7 million bases (30.93%) of the sequences discarded by the reference-guided assembly tool. The results indicate the robustness of LOCLAs contig detection algorithm on gap flanks. Furthermore, we show that 95% of the sequences filled in by LOCLA have over 80% accuracy compared with the HG002 reference genome published by the Human Pan-genome Reference Consortium. On the non-model organism, LOCLA enhanced the genome assembly of Bruguiera sexangula (JAHLGP000000000) by decreasing 41.4% of its gaps and raising the Benchmarking Universal Single-Copy Orthologs (BUSCO) analysis score to 98.10%. LOCLA can optimize de novo and reference-guided assemblies using varied sequencing reads. The final assemblies produced by LOCLA have improved in both quantity and quality. The increased gene content may provide a valuable resource in advancing personalized medicine.

genomics↗

bioRxiv2023

Temporal transcriptome analysis of the endothelial response to tumour necrosis factor

The vascular endothelium acts as a dynamic interface between blood and tissue and provides an anti-inflammatory and anti-thrombotic surface under normal conditions. Tumour necrosis factor- (TNF), a cytokine that drives acute and chronic inflammation, induces numerous transcriptional changes in endothelial cells (EC). However, the overall temporal dynamics of this response have not been fully elucidated. Here, we conducted an extended time-course analysis of the EC response to TNF, from 30 minutes to 72 hours. We identified TNF-regulated genes and used weighted gene correlation network analysis (WGCNA) to decipher co-expression profiles, uncovering two distinct temporal phases - an acute response initiated between 1-4 hours, followed by a later phase initiated between 12-24 hours. Several previously uncharacterised genes were strongly regulated during the acute phase, while the majority in the later phase were interferon-stimulated genes (ISGs). A lack of interferon transcription indicated that this TNF-induced ISG expression was independent of de novo interferon production and autocrine signalling. Furthermore, we observed two different groups of genes whose transcription was inhibited by TNF, those that resolved towards baseline levels over time, and those that did not. Our study provides insights into the global temporal dynamics of the EC transcriptional response to TNF, highlighting distinct gene expression patterns during the acute and later phases. These findings may be useful in understanding the role of EC in inflammation and developing TNF signalling-targeting therapies.

cell biology↗

bioRxiv2023

Embryo-specific epigenetic mechanisms reconstitute the CHH methylation landscape during Arabidopsis embryogenesis

The modification of flowering plant DNA by CHH methylation acts primarily to silence transposable elements, of which many active copies are present in Arabidopsis thaliana. During embryogenesis, the CHH methylation landscape is dramatically reprogrammed, resulting in exceedingly high levels of this modification upon mature embryo formation. The mechanisms constituting the remodeling process, and its function in embryos, are unclear. Here, we isolate embryos from Arabidopsis plants harboring mutations for key components of the pathways that confer CHH methylation, namely RNA-directed DNA methylation (RdDM) and the Chromomethylase 2 (CMT2) pathways. We reveal that embryos are more methylated than leaves at shared CMT2 and RdDM targeting loci, accounting for most embryonic CHH hypermethylation. While the majority of embryo CHH methylated loci overlap with those in somatic tissues, a subset of conventional pericentric CMT2-methylated loci are instead targeted by RdDM in embryos. These loci, termed embRdDM exhibit intermediate H3K9me2 levels, associated with increased chromatin accessibility. Strikingly, more than 50% of the embRdDM loci in pollen vegetative (nurse) cells and ddm1 mutant somatic tissues are also targeted by RdDM, and these tissues were also reported to exhibit increased chromatin accessibility in pericentric heterochromatin. Furthermore, the root columella stem cell niche also displays CHH hypermethylation and an enriched presence of small RNAs at embRdDM loci. Finally, we observe a significant overlap of CHH hypermethylated loci with endosperm DEMETER targeting sites, suggesting that non-cell autonomous communication within the seed may contribute to the epigenetic landscape of the embryo. However, similar overlap with vegetative cell DEMETER targets indicates that the chromatin landscape that allows DEMETER access is mirrored in developing embryos, permitting CHH methylation catalysis at the same loci. Our findings demonstrate that both conserved and embryo-specific epigenetic mechanisms reshape CHH methylation profiles in the dynamic chromatin environment of embryogenesis.

plant biology↗

bioRxiv2023

H2A.X mutants exhibit enhanced DNA demethylation in Arabidopsis thaliana

H2A.X is an H2A variant histone in eukaryotes, unique for its ability to respond to DNA damage, initiating the DNA repair pathway. H2A.X replacement within the histone octamer is mediated by the FAcilitates Chromatin Transactions (FACT) complex, a key chromatin remodeler. FACT is required for DEMETER (DME)-mediated DNA demethylation at certain loci in Arabidopsis thaliana female gametophytes during reproduction, though it is not known how FACT targets DME sites. Here, we investigated whether H2AX is involved in DME- and FACT-mediated DNA demethylation during Arabidopsis reproduction. We show that h2a.x mutants are more sensitive to genotoxic stress, consistent with previous reports. H2A.X fused to the Green Fluorescent Protein (GFP) gene under the H2A.X promoter was highly expressed in newly developing Arabidopsis tissues, including in male and female gametophytes. We examined DNA methylation in h2a.x developing seeds using whole genome bisulfite sequencing, and found that CG DNA methylation in the developing endosperm, but not the embryo, is decreased genome-wide in h2a.x mutants, predominately in transposons and intergenic DNA. Hypomethylated sites overlapped 62 % with canonical DME loci. These data indicate that H2A.X is not required for DME function, but is important for DNA methylation homeostasis in the unique chromatin environment of Arabidopsis endosperm.

plant biology↗

bioRxiv2022

Heterogeneity in the gene regulatory landscape of leiomyosarcoma.

Soft-tissue sarcomas are group of rare, tremendously heterogeneous, and highly aggressive malignancies. Characterizing inter-tumor heterogeneity is crucial for selecting suitable sarcoma therapy, as the presence of diverse molecular subgroups of patients can be associated with disease outcome or response to treatment. While cancer subtypes are often characterized by differences in gene expression, the mechanisms that drive these differences are generally unknown. We therefore set out to model the regulatory mechanisms driving sarcoma heterogeneity. We subtyped soft-tissue sarcomas based on patient-specific, genome-wide gene regulatory networks and found pronounced regulatory heterogeneity in leiomyosarcoma--one of the most common soft-tissue sarcomas subtypes that arises in smooth muscle tissue. To characterize this regulatory heterogeneity, we developed a new computational framework. This method, PORCUPINE, combines knowledge on biological pathways with permutation-based network analysis to identify pathways that exhibit significant regulatory heterogeneity across a patient population. We applied PORCUPINE to patient-specific leiomyosarcoma networks modeled on data from The Cancer Genome Atlas and validated our results in an independent dataset from the German Cancer Research Center. PORCUPINE identified 37 heterogeneously regulated pathways, including pathways that represent potential targets for treatment of subgroups of leiomyosarcoma patients, such as FGFR and CTLA4 inhibitory signaling. We validated the detected regulatory heterogeneity through analysis of networks and chromatin states in leiomyosarcoma cell lines. In addition, we showed that the heterogeneity identified with PORCUPINE is not associated with methylation profiles or clinical features, thereby suggesting an independent mechanism of patient heterogeneity driven by the complex landscape of gene regulatory interactions.

bioinformatics↗

bioRxiv2021

GABOLA: A Reliable Gap-Filling Strategy for de novo Chromosome-Level Assembly

We propose a novel method, GABOLA, which utilizes long-range genomic information provided by accurate linked short reads jointly with long reads to improve the integrity and resolution of whole genome assemblies especially in complex genetic regions. We validated GABOLA on human and Japanese eel genomes. On the two human samples, we filled in more bases spanning 23.3Mbp and 46.2Mbp than Supernova assembler, covering over 3,200 functional genes which includes 8,500 exons and 15,000 transcripts. Among them, multiple genes related to various types of cancer were identified. Moreover, we discovered additional 11,031,487 base pairs of repeat sequences and 218 exclusive repeat patterns, some of which are known to be linked to several disorders such as neuron degenerative diseases. As for the eel genome, we successfully raised the genetic benchmarking score to 94.6% while adding 24.7 million base pairs. These results manifest the capability of GABOLA in the optimization of whole genome assembly and the potential in precise disease diagnosis and high-quality non-model organism breeding. Availability: The docker image and source code of GABOLA assembler are available at https://hub.docker.com/r/lsbnb/gabola and https://github.com/lsbnb/gabola respectively.

genomics↗

bioRxiv2021

Adjustment of spurious correlations in co-expression measurements from RNA-Sequencing data

1Gene co-expression measurements are widely used in computational biology to identify coordinated expression patterns across a group of samples, which may indicate that these genes are controlled by the same transcriptional regulatory program, or involved in common biological processes. Gene co-expression is generally estimated from RNA-Sequencing data, which are commonly normalized to remove technical variability. Here, we demonstrate that certain normalization methods, in particular quantile-based methods, can introduce false-positive associations between genes, and that this can consequently hamper downstream co-expression network analysis. Quantile-based normalization can, however, be extremely powerful. In particular when preprocessing large-scale heterogeneous data, quantile-based normalization methods such as smooth quantile normalization can be applied to remove technical variability while maintaining global differences in expression for samples with different biological attributes. We therefore developed SNAIL, a normalization method based on smooth quantile normalization specifically designed for modeling of co-expression measurements. We show that SNAIL avoids formation of false-positive associations in co-expression as well as in downstream network analyses. Using SNAIL, one can avoid arbitrary gene filtering and retain associations to genes that only express in small subgroups of samples. This highlights the methods potential future impact on network modeling and other association-based approaches in large-scale heterogeneous data.

bioinformatics↗

bioRxiv2021

immuneML: an ecosystem for machine learning analysis of adaptive immune receptor repertoires

Adaptive immune receptor repertoires (AIRR) are key targets for biomedical research as they record past and ongoing adaptive immune responses. The capacity of machine learning (ML) to identify complex discriminative sequence patterns renders it an ideal approach for AIRR-based diagnostic and therapeutic discovery. To date, widespread adoption of AIRR ML has been inhibited by a lack of reproducibility, transparency, and interoperability. immuneML (immuneml.uio.no) addresses these concerns by implementing each step of the AIRR ML process in an extensible, open-source software ecosystem that is based on fully specified and shareable workflows. To facilitate widespread user adoption, immuneML is available as a command-line tool and through an intuitive Galaxy web interface, and extensive documentation of workflows is provided. We demonstrate the broad applicability of immuneML by (i) reproducing a large-scale study on immune state prediction, (ii) developing, integrating, and applying a novel method for antigen specificity prediction, and (iii) showcasing streamlined interpretability-focused benchmarking of AIRR ML.

bioinformatics↗

bioRxiv2018

GenEpi: Gene-based Epistasis Discovery Using Machine Learning

Genome-wide association studies (GWAS) provide a powerful means to identify associations between genetic variants and phenotypes. However, GWAS techniques for detecting epistasis, the interactions between genetic variants associated with phenotypes, are still limited. We believe that developing an efficient and effective GWAS method to detect epistasis will be a key for discovering sophisticated pathogenesis, which is especially important for complex diseases such as Alzheimers disease (AD). In this regard, this study presents GenEpi, a computational package to uncover epistasis associated with phenotypes by the proposed machine learning approach. GenEpi identifies both within-gene and cross-gene epistasis through a two-stage modeling workflow. In both stages, GenEpi adopts two-element combinatorial encoding when producing features and constructs the prediction models by L1-regularized regression with stability selection. The simulated data showed that GenEpi outperforms other widely-used methods on detecting ground-truth epistasis. As real data is concerned, this study uses AD as an example to reveal the capability of GenEpi in finding disease-related variants and variant interactions that show both biological meanings and predictive power. Availability: GenEpi is an open-source python package and available free of charge only for non-commercial users. The package can be downloaded from https://github.com/Chester75321/GenEpi, and has also been published on The Python Package Index.

bioinformatics↗

bioRxiv2018

Effect of de novo transcriptome assembly on transcript quantification

BackgroundCorrect quantification of transcript expression is essential to understand the functional products of the genome in different physiological conditions and developmental stages. Recently, the development of high-throughput RNA sequencing (RNA-Seq) allows the researchers to perform transcriptome analysis for the organisms without the reference genome and transcriptome. For such projects, de novo transcriptome assembly must be carried out prior to quantification. However, a large number of erroneous contigs produced by the assemblers might result in unreliable estimation on the abundance of transcripts. In this regard, this study comprehensively investigates how assembly quality affects the performance of quantification for RNA-Seq analysis based on de novo transcriptome assembly.\n\nResultsSeveral important factors that might seriously affect the accuracy of the RNA-Seq analysis were thoroughly discussed. First, we found that the assemblers perform comparatively well for the transcriptomes with lower biological complexity. Second, we examined the over-extended and incomplete contigs, and then demonstrated that assembly completeness has a strong impact on the estimation of contig abundance. Lastly, we investigated the behavior of the quantifiers with respect to sequence ambiguity which might be originally present in the transcriptome or accidentally produced by assemblers. The results suggest that the quantifiers often over-estimate the expression of family-collapse contigs and under-estimate the expression of duplicated contigs. For organisms without reference transcriptome, it remained challenging to detect the inaccurate abundance estimation on family-collapse contigs. On the contrary, we observed that the situation of under-estimation on duplicated contigs can be warned through analyzing the read distribution of the duplicated contigs.\n\nConclusionsIn summary, we explicated the behavior of quantifiers when erroneous contigs are present and we outlined the potential problems that the assemblers might cause for the downstream analysis of RNA-Seq. We anticipate the analytic results conducted in this study provides valuable insights for future development of transcriptome assembly and quantification.\n\nAvailabilitywe proposed an open-source Python based package QuantEval that builds connected components for the assembled contigs based on sequence similarity and evaluates the quantification results for each connected component. The package can be downloaded from https://github.com/dn070017/QuantEval.

bioinformatics↗

bioRxiv2017

The FACT complex is required for DNA demethylation at heterochromatin during reproduction in Arabidopsis

The DEMETER (DME) DNA glycosylase catalyzes genome-wide DNA demethylation and is required for endosperm genomic imprinting and embryo viability. Targets of DME-mediated DNA demethylation reside in small, euchromatic, AT-rich transposons and at the boundaries of large transposons, but how DME interacts with these diverse chromatin states is unknown. The STRUCTURE SPECIFIC RECOGNITION PROTEIN 1 (SSRP1), subunit of the chromatin remodeler FAcilitates Chromatin Transactions (FACT), was previously shown to be involved in the DME-dependent regulation of genomic imprinting in Arabidopsis endosperm. Therefore, to investigate the interaction between DME and chromatin, we focused on the activity of the two FACT subunits, SSRP1 and SUPPRESSOR of TY16 (SPT16), during reproduction in Arabidopsis. We find that FACT co-localizes with nuclear DME in vivo, and that DME has two classes of target sites, the first being euchromatic and accessible to DME, but the second, representing over half of DME targets, requiring the action of FACT for DME-mediated DNA demethylation genome-wide. Our results show that the FACT-dependent DME targets are GC-rich heterochromatin domains with high nucleosome occupancy enriched with H3K9me2 and H3K27me1. Further, we demonstrate that heterochromatin-associated linker histone H1 specifically mediates the requirement for FACT at a subset of DME-target loci. Overall, our results demonstrate that FACT is required for DME targeting by facilitating its access to heterochromatin.

plant biology↗