Biology subjects

Ho, J. W. K.

Publications and source records attributed to Ho, J. W. K..

At least 19 recordsLinked to original sources

bioRxiv2026

CLEAR-ST: Physics-informed probabilistic decontamination of spatial transcriptomics by modeling mRNA lateral diffusion

Spatial transcriptomics is a rapidly evolving technology that allows for the measurement of gene expression in a spatially resolved manner. However, one technical problem that occurs for many sequencing-based spatial transcriptomics platforms is the presence of mRNA lateral diffusion, where mRNA from one spot can bind to probes in another spot, leading to contamination and inaccurate gene expression measurements. In Visium-like assays, this artifact is often visible as structured out-of-tissue signal and boundary-associated expression halos, yet its magnitude, spatial decay, and directional bias vary substantially across samples. Here, we present CLEAR-ST, a physics-informed probabilistic framework for correcting diffusion-like contamination in spatial transcriptomics data. CLEAR-ST infers a latent clean expression field using a denoising autoencoder and links it to the observed counts through a graph-Laplacian forward contamination model with learnable diffusion parameters, finally evaluated with a selectable count likelihood. We first conducted a comprehensive comparison between 10X official and independently generated Visium samples, demonstrating that out-of-tissue count profiles are highly related to nearby in-tissue expression, more concentrated near tissue boundaries, and diffusion directions across genes are likely coherent. Across real samples with varying contamination burden, CLEAR-ST improved spatial domain recovery, increased gene-level spatial autocorrelation, and enhanced the biological specificity of downstream analyses such as marker gene discovery, pathway identification and cell type deconvolution. Compared to benchmark methods, CLEAR-ST showed consistent gains in clustering quality and concordance with manual annotations. Together, CLEAR-ST provides an interpretable and practical approach for diffusion-aware correction of capture-based spatial transcriptomics data.

bioinformatics↗

bioRxiv2026

vOMIX-MEGA: An ultra-fast end-to-end pipeline for terabyte-scale viral metagenomics analysis.

Viral identification for terabyte-scale metagenomic data is limited by scalability and computational resources. We present vOMIX-MEGA, an end-to-end viral metagenomic framework that overcomes performance bottlenecks by significantly improving parallelization and memory usage in critical steps. Benchmarked on empirical datasets, it completes processing in up to 50 minutes with 24 GB of RAM, bypassing four other state-of-the-art pipelines that require 7 hours (383 GB) to 14 days (32 GB). vOMIX-MEGA is on average 21% and 13% more accurate when benchmarked on mock and experimental data and is available via https://github.com/holab-hku/vOMIX-MEGA.

bioinformatics↗

bioRxiv2026

Spatial mitochondrial lineage tracing uncovers a premetastatic niche and microenvironment programmed fate switching in osteosarcoma

Osteosarcoma progression involves complex spatiotemporal dynamics, yet the lineage relationships underlying tumor cell fate decisions remain poorly understood. Here we integrate single cell and spatial transcriptomics and mitochondrial variants-based lineage tracing to map the evolution of a late-stage, non-metastatic osteosarcoma. Single cell pseudotemporal reconstruction of malignant cells reveals a stepwise cascade of transcriptional programs that instructs bifurcation of COL3A1 progenitors into ALPL osteoblastic and THY1 mesenchymal lineages. Critically, we demonstrate that the metastatic THY1 mesenchymal cell fate is not governed by cell-autonomous transcriptional programs alone; this pro-tumorigenic state requires strict spatial licensing via direct co-localization with PLVAP dysfunctional endothelia to form a pro-metastatic niche (THY1_Endo) enriched for PTN/NOTCH signaling. To distinguish clonal ancestry from microenvironmental plasticity, we leveraged spatially resolved somatic mitochondrial variants as endogenous lineage tracers. A spatial autocorrelation filter confirmed significant clonal spatial coherence (Z = 10.07, p {approx} 4x10-{superscript 2}). Phylogenetic reconstruction using a Wright-Fisher drift model coupled with a hidden Markov tree (WF-HMT) revealed a striking directional transition rate from COL3A1 to THY1 tumor cells (724.08) versus the reverse (4.09)--a [~]177-fold asymmetry--whereas the COL3A1 to ALPL transition rate was lower (0.47). Notably, drift-agnostic coarse VAF clustering failed to resolve ALPL versus THY1 bifurcation, underscoring that explicit modeling of stochastic mitochondrial drift is essential for uncovering directional fate locking. We propose a "time-space-lineage" relay model where sequential microenvironmental signals program tumor cell fate and pre-metastatic niche assembly through irreversible clonal commitment. This multi-dimensional framework bridges single-cell states, spatial organization, temporal dynamics, and clonal ancestry --underscoring the power of somatic mitochondrial variants to resolve clonal history and intercept cancer evolution.

bioinformatics↗

bioRxiv2025

Associations Between Gut Microbiome and 24-Hour Blood Pressure Variability: A Cross-sectional Study Highlighting Sex Differences and Potential Therapeutic Targets

Blood pressure (BP) variability is an independent risk factor for cardiovascular disease (CVD). While gut microbiota (GM) and GM-derived short-chain fatty acids (SCFAs) are recognized to play a key role in BP regulation, their association with BP variability remains poorly understood. This cross-sectional study of 241 community-dwelling individuals from Hong Kong (113 men and 128 women, mean age 54{+/-}6 years) without symptomatic CVD examined the associations of GM, characterized by shotgun sequencing, and plasma SCFAs, with systolic and diastolic blood pressure (SBP/DBP) variability assessed by 24-hour BP monitoring. GM analyses, using covariate-adjusted statistical models, revealed that higher 24-hour SBP CoV associated negatively with GM -diversity (Shannon and Simpsons index, P<0.05) and positively with Firmicutes/Bacteroidetes ratio, driven by the female cohort. Parabacteroides merdae, Bacteroides dorei, Bifidobacterium pseudocatenulatum, Alistipes finegoldii and Bacteroides intestinalis were negatively associated with indices of SBP/DBP variability in a sex-specific manner. Further analysis indicated that B. dorei may mediate SBP CoV via plasma iso-butyric acid in women (bootstrapping 95% CI: -3.6 to -0.19; P<0.05). We demonstrated that higher SBP variability is associated with markers of gut dysbiosis and a reduction in beneficial gut bacteria, particularly in women. Notably, we identified several gut bacterial species with potential therapeutic implications for managing 24-hour SBP/DBP variability, warranting further investigation.

microbiology↗

bioRxiv2025

Can large language models reliably extract human disease genes from full-text scientific literature?

Manual extraction of high-fidelity gene-disease-phenotype information from human genetics literature is a labor-intensive task that requires trained human genetics researchers to read through many primary research papers. This presents a major challenge for maintaining up-to-date human disease genetic databases. Recent exploration into large language models (LLMs) opens new directions in automating this manual process. However, most approaches depend on pre-training, finetuning, or specialized generative artificial intelligence (GenAI) tools, but there is a lack of empirical evidence to show whether commercially-available LLMs can be directly used to reliably extract gene-disease-phenotype for human genetic diseases. Herein, we perform a benchmark of the use of three zero-shot prompted LLMs, namely GPT-4, DeepSeek and Claude, without task-specific fine-tuning, to extract human genetic information directly from full text of scientific papers. Using known congenital heart diseases (CHD) genes found in the open access CHDgene database (https://chdgene.victorchang.edu.au/) as the benchmark data set, GPT-4o achieved overall 88.8% extraction accuracy across 23 gene entries containing over 57 references, with 100% accuracy in gene name, 78.3% and 76.7% in disease and phenotype fields respectively. This work introduces a lightweight, easy-to-deploy, and yet robust LLM-based agent named GeneAgent, analyze sources of disagreement, and highlight the feasibility of integrating powerful LLM into genetic evidence synthesis workflows. Highlight- First systematic benchmark of LLMs for extracting human gene-disease-phenotype relationships from full-text biomedical articles - GeneAgent: a lightweight, highly accurate prompt-only LLM agent - New domain task-specific evaluation framework

bioinformatics↗

bioRxiv2025

AtlasAgent: Vision language model and Agent-guided Framework for Evaluation of Atlas-scale Single-cell Integration

As single-cell omics transitions into the era of AI-virtual cells (AIVC), where large-scale single-cell data integration becomes prevalent, the computational demands of integration evaluation emerge as critical scalability bottlenecks. Traditional integration evaluation pipelines, requiring metrics like k-nearest-neighbor batch effect test (kBET) and Local Inverse Simpsons Index (iLISI) employed by state-of-the-art scIB method, often demand large computational resources and long runtimes, making them infeasible for large scale integration studies. Herein, we present AtlasAgent, the first vision-language model (VLM)-powered and AI agent framework to accelerate atlas-scale integration evaluation at unprecedented speed and scale. We systematically evaluate batch correction quality, biological signal preservation and overcorrection risks using chain-of-thought reasoning in conjunction with few-shot and zero-shot prompting strategies. AtlasAgent completes evaluation within 32 seconds, in contrast to scIBs runtime of 5.55 hours in GPU, while identifying the scIB-determint best integration methods within the top-3 in 88.3% of the time, lowering evaluation time from hours to seconds while preserving alignment with domain expert reasoning. AtlasAgent pioneers the use of VLMs to realize scalable and rapid integration evaluation at atlas scale.

bioinformatics↗

bioRxiv2024

Accurate prediction of CRISPR editing outcomes in somatic cell lines and zygote with few-shot learning

The CRISPR-Cas system has revolutionized gene editing, while its outcome prediction remains unsatisfactory, especially in new cell states due to their distinct DNA repair preferences. In this study, we introduce inDecay, a flexible system for predicting CRISPR editing outcomes from target sequence, returning probabilities of nearly the full spectrum of indel events. Uniquely, inDecay utilizes informative and parameter-efficient features for each indel event and incorporates cell-type-specific repair preferences through a multi-stage design. While both inDecay and existing methods achieve accurate results for prediction within cell lines, only inDecay with transfer learning can retain the high performance for cross-cell line prediction. We then applied inDecay to mouse embryo editing using our newly generated data and observed remarkable accuracy by including as few as 30 fine-tuning embryonic samples. Notably, inDecay is the first software to predict embryonic editing. Therefore, our few-shot learning-supported system may accelerate guide RNA prioritization in mouse model generation, mammal embryonic gene editing, and cellular therapeutics.

bioinformatics↗

bioRxiv2024

GeneRAG: Enhancing Large Language Models with Gene-Related Task by Retrieval-Augmented Generation

Large Language Models (LLMs) like GPT-4 have revolutionized natural language processing and are used in gene analysis, but their gene knowledge is incomplete. Fine-tuning LLMs with external data is costly and resource-intensive. Retrieval-Augmented Generation (RAG) integrates relevant external information dynamically. We introduce GO_SCPLOWENEC_SCPLOWRAG, a frame-work that enhances LLMs gene-related capabilities using RAG and the Maximal Marginal Relevance (MMR) algorithm. Evaluations with datasets from the National Center for Biotechnology Information (NCBI) show that GO_SCPLOWENEC_SCPLOWRAG outperforms GPT-3.5 and GPT-4, with a 39% improvement in answering gene questions, a 43% performance increase in cell type annotation, and a 0.25 decrease in error rates for gene interaction prediction. These results highlight GO_SCPLOWENEC_SCPLOWRAGs potential to bridge a critical gap in LLM capabilities for more effective applications in genetics.

bioinformatics↗

bioRxiv2024

Metagenomic coverage bias at transcription start sites is correlated with gene expression

Metagenomic sequencing is presumed to provide unbiased sampling of all the genetic material in a sample. Downstream analysis methods, such as binning, gene copy number analysis, structural variations, or single nucleotide polymorphism analysis, commonly assume an even distribution across the genome after accounting for known artefacts such as GC content. We discovered coverage bias across gut microbiome species, manifesting as a difference in coverage before and after bacterial transcription start sites. Using matched metatranscriptomic and metagenomic sequencing data, we demonstrate that this bias correlates with gene expression. Potential artefacts such as the sequencing technology, reference genome used for alignment, and mappability bias were investigated across multiple datasets and shown to not be factors for association. While GC bias was found correlated with coverage bias, the association of coverage bias with gene expression remains significant after adjusting for GC bias. Paired-end read mapping demonstrated an enrichment in 5 read ends immediately downstream of the TSS which was partly a byproduct of unmapped reads upstream of the TSS. Our observations suggest the existence of strain-level variation where sequence variation in the promoter site region is preventing proper read alignment to the reference genome. The correlation of this phenomenon with gene expression may also reflect evolutionary footprints for fine-tuning the regulation of gene expression. Understanding the source of this sequence variation and the biological implications of this artefact will be useful not only to better characterise microbial functions but also to improve interpretations of strain level dynamics. ImportanceSequencing coverage calculated from metagenomic sequencing data is extensively used in the microbiome field, providing valuable information about microbial abundances, gene (functional) abundances, growth rates, and genomic variations. Understanding factors that impact the distribution of coverage along genomes is therefore important for multiple applications. In this study, we report on uneven read coverage across the transcription start sites of bacterial genomes that is correlated with gene expression levels. We determine that this bias is independent of multiple factors including GC bias, and arises due to higher strain divergence from reference genomes upstream of the transcript start site. We propose that evolutionary finetuning of gene expression in competitive microbial ecosystems can drive genetic mutations at the promoter site. Our findings suggest the potential to glean gene regulatory information from metagenomic data, and better understand how ecological factors shape genomes in the microbiome and their sequencing coverage.

bioinformatics↗

bioRxiv2024

StaVia: Spatially and temporally aware cartography with higher order random walks for cell atlases

Single-cell atlases are critical for unraveling the cellular basis of health and disease, yet their sheer diversity and vast data landscape in time and space pose daunting computational challenges pertaining to the delineation of multiple simultaneously emerging lineages, the integration of spatial and temporal information, and the visualization of trajectories across large atlases with enough resolution to observe localized transitions. To tackle this intricacy, we introduce StaVia, a computational framework that synergizes multi-faceted single-cell data--spanning time-series data, spatial gene expression patterns, and directional trends from RNA velocity-- with higher-order random walks that leverage the memory of cells past states. StaVia fuses this method with a cartographic "Atlas View" that offers intuitive graph visualization, simultaneously capturing the nuanced details of cellular development at single-cell resolution as well as the broader connectivity of cell lineages, avoiding common pitfalls of merged distinct trajectories or missed transitional states seen in existing methods which are all memoryless. Notably, we demonstrate that StaVia unlocks new insights into placode development, radial glia pluripotency during neurulation, and the transitions pivotal to these processes in a large-scale Zebrafish developmental atlas. StaVia also allows spatially aware cartography that captures relationships between cell populations based on their spatial location as well as their gene expressions in a MERFISH dataset - underscoring its potential to dissect complex biological landscapes in both spatial and temporal contexts.

bioinformatics↗

bioRxiv2023

MetaQuad: Shared Informative Variants Discovery in Metagenomic Samples

MotivationStrain-level analysis of metagenomic data has garnered significant interest in recent years. Microbial single nucleotide polymorphisms (SNPs) are genomic variants that can reflect strain-level differences within a microbial species. The diversity and emergence of SNPs in microbial genomes may reveal evolutionary history and environmental adaptation in microbial populations. However, efficient discovery of shared polymorphic variants in a large collection metagenomic samples remains a computational challenge. ResultsMetaQuad employs a density-based clustering technique to efficiently differentiate shared variants from non-polymorphic sites using shotgun metagenomic data. Empirical comparisons with other state-of-the-art methods show that MetaQuad significantly reduces the number of false-positive SNPs without greatly affecting the true-positive rate. We used MetaQuad to identify antibiotic-associated variants in patients who underwent Helicobacter pylori eradication therapy. MetaQuad detected 7,591 variants across 529 antibiotic resistance genes. The nucleotide diversity of some genes is increased six weeks after antibiotic treatment, potentially indicating the role of these genes in specific antibiotic treatments. AvailabilityMetaQuad is an open-source Python package available via https://github.com/holab-hku/MetaQuad. Contactjwkho@hku.hk Supplementary informationSupplementary data are available at XXXX online.

genomics↗

bioRxiv2023

Single-cell mitochondrial variant enrichment resolved clonal tracking and spatial architecture in human embryonic hematopoiesis

The ability to perform lineage tracing at the single-cell level is critical to reconstructing dynamic transitions during cell differentiation. However, prospective tracing approaches inevitably encounter outstanding challenges including barcoding precision, barcode diversity, and detection efficiency, which can skew inferred lineage relationships. Human pluripotent stem cells (hPSCs) even face risks of DNA-damage-induced toxicity-related cell death. We explored the use of naturally occurring somatic mutations in mitochondrial transcripts detected in single-cell RNA-seq as genetic lineage barcodes in hPSCs. In this study, we used an enrichment of scRNA-seq mitochondrial reads and a robust computational method to identify clonally relevant mitochondrial variants as endogenous genetic barcodes for clonal tracking of early embryonic hematopoiesis from hPSC. We modeled the development of embryonic tissues from hPSCs and delineated cell fate specification by integrating synthetic barcoding with mitochondrial lineage tracing. Using a biophysical model, we reconstructed the sequential transcriptional logic of fate specification and its underlying regulatory network. We further applied mitochondrial lineage tracing to spatial transcriptomics, which enabled us to identify the spatial clonal architecture of human embryonic organoids. Our analysis revealed that this spatial zonation was orchestrated by NOTCH-mediated crosstalk between stromal cells and hematopoietic progenitors. Our multi-modal framework links clonal dynamics with niche-specific fate decisions, providing a generalizable approach to dissecting tissue organization in development and disease. This study underscores the utility of mitochondrial variants as endogenous markers for high-resolution spatial clonal tracking in stem cell-derived organoid models of human development.

developmental biology↗

bioRxiv2023

Vulture: Cloud-enabled scalable mining of microbial reads in public scRNA-seq data

The rapidly growing collection of public single-cell sequencing data have become a valuable resource for molecular, cellular and microbial discovery. Previous studies mostly overlooked detecting pathogens in human single-cell sequencing data. Moreover, existing bioinformatics tools lack the scalability to deal with big public data. We introduce Vulture, a scalable cloud-based pipeline that performs microbial calling for single-cell RNA sequencing (scRNA-seq) data, enabling meta-analysis of host-microbial studies from the public domain. In our scalability benchmarking experiments, Vulture can outperform the state-of-the-art cloud-based pipeline Cumulus with a 40% and 80% reduction of runtime and cost, respectively. Furthermore, Vulture is 2-10 times faster than PathogenTrack and Venus, while generating comparable results. We applied Vulture to two COVID-19, three hepatocellular carcinoma (HCC), and two gastric cancer human patient cohorts with public sequencing reads data from scRNA-seq experiments and discovered cell-type specific enrichment of SARS-CoV2, hepatitis B virus (HBV), and H. pylori positive cells, respectively. In the HCC analysis, all cohorts showed hepatocyte-only enrichment of HBV, with cell subtype-associated HBV enrichment based on inferred copy number variations. In summary, Vulture presents a scalable and economical framework to mine unknown host-microbial interactions from large-scale public scRNA-seq data. Vulture is available via an open-source license at https://github.com/holab-hku/Vulture.

bioinformatics↗

bioRxiv2022

PD-L1 regulates inflammatory macrophage development from human pluripotent stem cells by maintaining interferon-gamma signal

PD-L1 (programmed death-ligand 1) serves as a pivotal immune checkpoint in both the innate and adaptive immune systems. PD-L1 is expressed in macrophages in response to interferon-gamma (IFN{gamma}). We examined whether PD-L1 might regulate macrophage development. We established PD-L1-/- human pluripotent stem cells, differentiated them into macrophages, and observed a 60% reduction of CD11B+CD45+ macrophages in PD-L1-/-, orthogonally verified with PD-L1 inhibitor BMS-1166 reduced macrophages to the same fold. Single-cell RNA sequencing further confirmed the 60% reduction of macrophages as well as the down-regulation of macrophage-defining transcription factors SPI1, KLF6, and MAFB. Further, PD-L1-/- macrophages reduced the level of inflammatory signals such as NF{kappa}B, TNF, and chemokines CXCL and CCL families. Whilst anti-inflammatory TGF-{beta} was upregulated. Finally, we identified that PD-L1-/- macrophages significantly down-regulated interferon-stimulated genes (ISGs) despite IFN{gamma} in differentiation media. Mechanistically, PD-L1-/- macrophages reduced IFNGR1 expression explaining that cells could not respond to IFN{gamma}. These data suggest that PD-L1 regulates inflammatory macrophage development by maintaining the IFN{gamma} signal.

immunology↗

bioRxiv2022

Covidscope: An atlas-scale COVID-19 resource for single-cell meta analysis at sample and cell levels

With the recent advancement in single-cell technologies and the increased availability of integrative tools, challenges arise in easy and fast access to large collections of cell atlas. Existing cell atlas portals rarely are open sourced and adaptable, and do not support meta-analysis at cell level. Here, we present an open source, highly optimised and scalable architecture, named Scope+, to allow quick access, meta-analysis and cell-level selection of the atlas data. We applied this architecture to our well-curated 5 million Covid-19 blood and immune cells, as a portal, Covidscope (https://covidsc.d24h.hk/). We achieved efficient access to atlas-scale data via three strategies, such as server-side rendering, novel database optimization strategies and an innovative architectural design. Scope+ serves as an open source architecture for researchers to build on with their own atlas, and demonstrated its capability in the Covidscope portal for an effective meta-analysis to atlas data at cellular resolution for reproducible research.

bioinformatics↗

bioRxiv2022

Inferring cell diversity in single cell data using consortium-scale epigenetic data as a biological anchor for cell identity

Methods for cell clustering and gene expression from single-cell RNA sequencing (scRNA-seq) data are essential for biological interpretation of cell processes. Here we present TRIAGE-Cluster which uses genome-wide epigenetic data from diverse bio-samples to identify genes demarcating cell diversity in scRNA-seq data. TRIAGE-Cluster integrates patterns of repressive chromatin deposited across diverse cell types with weighted density estimation to determine cell type clusters in a 2D UMAP space. We then present TRIAGE-ParseR, a machine learning method that evaluates gene expression rank lists to define gene groups governing the identity and function of cell types. We demonstrate the utility of this two-step approach using atlases of in vivo and in vitro cell diversification and organogenesis. We also provide a web accessible dashboard for analysis and download of data and software. Collectively, genome-wide epigenetic repression provides a versatile strategy to define cell diversity and study gene regulation of scRNA-seq data.

bioinformatics↗

bioRxiv2022

An integrated cell barcoding and computational analysis pipeline for scalable analysis of differentiation at single-cell resolution

Pluripotent stem cells provide a scalable approach to analyse molecular regulation of cell differentiation across multiple developmental lineage trajectories. In this study, we engineered barcoded iPSCs to generate an atlas of multilineage differentiation from pluripotency, encompassing a time-course of WNT-induced differentiation perturbed using modulators of WNT, BMP, and VEGF signalling. Computational mapping of in vitro cell types to in vivo developmental lineages revealed a diversity of iPSC-derived cell types comprising mesendoderm lineage cell types including lateral plate and paraxial mesoderm, neural crest, and primitive gut. Coupling this atlas of in vitro differentiation with Summary data-based Mendelian Randomisation analysis of human complex traits, we identify the WNT-inhibitor protein TMEM88 as a putative regulator of mesendodermal cell types governing development of diverse cardiovascular and anthropometric traits. Using genetic loss of function models, we show that TMEM88 is required for differentiation of diverse endoderm and mesoderm cell lineages in vitro and that TMEM88 knockout in vivo results in a significant dysregulation of arterial blood pressure. This study provides an atlas of multilineage iPSC differentiation coupled with new molecular, computational, and statistical genetic tools to dissect genetic determinants of mammalian developmental physiology.

genomics↗

bioRxiv2022

Multi-task learning uncovers robust translation cis-regulatory features

Many studies have found that sequence in the 5 untranslated regions (UTRs) impacts the translation rate of an mRNA, but the regulatory grammar that underpins this translation regulation remains elusive. Deep learning methods deployed to analyse massive sequencing datasets offer new solutions to motif discovery. However, existing works focused on extracting sequence motifs in individual datasets, which may not be generalisable to other datasets from the same cell type. We hypothesise that motifs that are genuinely involved in controlling translation rate are the ones that can be extracted from diverse datasets generated by different experimental techniques. In order to reveal more generalised cis-regulatory motifs for RNA translation, we develop a multi-task translation rate predictor, MTtrans, to integrate information from multiple datasets. Compared to single-task models, MTtrans reaches a higher prediction accuracy in all the benchmarked datasets generated by various experimental techniques. We show that features learnt in human samples are directly transferable to another dataset in yeast systems, demonstrating its robustness in identifying evolutionarily conserved sequence motifs. Furthermore, our newly generated experimental data corroborated the effect of most of the identified motifs based on MTtrans trained using multiple public datasets, further demonstrating the utility of MTtrans for discovering generalisable motifs. MTtrans effectively integrates biological insights from diverse experiments and allows robust extraction of translation-associated sequence motifs in 5UTR.

bioinformatics↗