Biology subjects

Chen, H.-L.

Publications and source records attributed to Chen, H.-L..

6 recordsLinked to original sources

bioRxiv2025

Template-based RNA structure prediction advanced through a blind code competition

Automatically predicting RNA 3D structure from sequence remains an unsolved challenge in biology and biotechnology. Here, we describe a Kaggle code competition engaging over 1700 teams and 43 previously unreleased structures to tackle this challenge. The top three submitted algorithms achieved scores within statistical error of the winners of the recent CASP16 competition. Unexpectedly, the top Kaggle strategy involved a pipeline for discovering 3D templates, without the use of deep learning. We integrated this template-modeling pipeline and other Kaggle strategies to develop a single model RNAPro that retrospectively outperformed individual Kaggle models on the same test set. These results suggest a growing importance of template-based modeling in RNA structure prediction.

biophysics↗

bioRxiv2025

Phenotypic and functional characterization of tumor-reactive T cells in malignant pleural effusions

BackgroundAdoptive cell therapy using tumor-infiltrating lymphocytes (TIL) is approved for the treatment of advanced melanoma but is limited by the need for patients to undergo surgical tumor resection. Malignant pleural effusions (MPE) may represent a more accessible source of tumor-reactive T cells. Here, we characterize the cellular composition as well as the transcriptional and functional properties of T cells from MPE compared with pulmonary metastasis and blood from a patient with melanoma. MethodsThe immune cellular composition was immunophenotyped by high-dimensional flow cytometry from synchronously collected MPE, a lung metastasis, and blood from a patient with metastatic melanoma. Sorted CD3+ T cells were profiled by single-cell RNA sequencing (scRNA-seq) and T cell receptor sequencing (scTCR-seq). TCR reactivity to autologous tumor was evaluated through in vitro activation assays with TCR-transduced Jurkat and autologous cancer cells. The killing capacity of ex vivo expanded T cells of autologous cancer cells was assessed through in vitro cytotoxicity assays. ResultsMPE had higher proportions of CD45+ immune cells and CD3+ T cells (70.5% vs 50%) compared with tumor and was enriched for effector CD8+ T cells, CCR7-CD45RA- effector memory CD4+ T cells, and quiescent CD25highCD127low regulatory CD4+ T cells. MPE T cells exhibited lower levels of co-inhibitory receptors (PD-1, LAG-3, TIGIT, TIM-3) expression relative to tumor. ScRNA-seq showed enrichment of NK-like effector CD8+ T cells in MPE. Pseudotime analysis indicated that MPE T cells were less exhausted than tumor T cells. The clonal repertoire of MPE and tumor highly overlapped, including 62.2% of predicted neoantigen-specific (NeoTCR) clonotypes. Notably, clonally-related NeoTCR T cells in MPE exhibited higher cytotoxic and stemness, and lower exhaustion signatures compared with sister clones in the tumor. Two of four selected NeoTCR clonotypes transduced in Jurkat cells demonstrated MHC class I-restricted reactivity in co-culture with autologous cancer cells. MPE T cells also readily expanded in the presence of high-dose IL-2 and demonstrated MHC class I-dependent killing of autologous cancer cells. ConclusionsMPE harbors polyclonal, tumor-reactive T cells with lower features of terminal exhaustion and higher cytotoxic potential relative to tumor T cells. MPE may therefore serve as a more accessible source for TIL therapy.

immunology↗

bioRxiv2024

Telomere-to-telomere sheep genome assembly reveals new variants associated with wool fineness trait

Ongoing efforts to improve sheep reference genome assemblies still leave many gaps and incomplete regions, resulting in a few common failures and errors in sheep genomic studies. Here, we report a complete, gap-free telomere-to-telomere (T2T) genome of a ram (T2T-sheep1.0) with a size of 2.85 Gb, including all autosomes and chromosomes X and Y. It adds 220.05 Mb of previously unresolved regions (PURs) and 754 new genes to the most updated reference assembly, ARS-UI_Ramb_v3.0, and contains four types of repeat units (SatI, SatII, SatIII, and CenY) in the centromeric regions. T2T-sheep1.0 exhibits a base accuracy of >99.999%, corrects several structural errors in previous reference assemblies, and improves structural variant (SV) detection in repetitive sequences. We identified 192,265 SVs, including 16,885 new SVs in the PURs, from the PacBio long-read sequences of 18 global representative sheep. With the whole-genome short-read sequences of 810 wild and domestic sheep representing 158 global populations and seven wild species, the use of T2T-sheep1.0 as the reference genome has improved population genetic analysis based on [~]133.31 million SNPs and 1,265,266 SVs, including 2,664,979 novel SNPs and 196,471 novel SVs. T2T-sheep1.0 improves selective tests by detecting several novel genes and variants, including those associated with domestication (e.g., ABCC4) and selection for the wool fineness trait (e.g., FOXQ1) in tandemly duplicated regions.

genomics↗

bioRxiv2024

Telomere-to-telomere genome assembly of a male goat reveals novel variants associated with cashmere traits

A complete goat (Capra hircus) reference genome enhances analyses of genetic variation, thus providing novel insights into domestication and selection in goats and related species. Here, we assembled a telomere-to-telomere (T2T) gap-free genome (2.86 Gb) from a cashmere goat (T2T-goat1.0), which included a Y chromosome of 20.9 Mb. With a base accuracy of >99.999%, T2T-goat1.0 corrected numerous genome-wide structural and base errors in previous assemblies and added 288.5 Mb of previously unresolved regions (PURs) and 446 newly assembled genes to the reference genome. We sequenced the genomes of five representative goat breeds for PacBio reads, and used T2T-goat1.0 as a reference to identify a total of 63,417 structural variations (SVs) with up to 4711 SVs (7.41%) in the PURs and 507 overlapping genes significantly enriched (Padj. < 0.05) in the keratin filament pathway. T2T-goat1.0 was applied in population analyses of global wild and domestic goats, which revealed 32,419 SVs and 25,397,794 SNPs, including 870 SVs and 545,026 SNPs in the PURs. Also, our analyses revealed a set of novel selective variants and genes associated with domestication (e.g., NKG2D and ABCC4) and cashmere traits (e.g., ABCC4 and ASIP).

genomics↗

bioRxiv2021

Homo-harringtonine (HHT) - A highly effective drug against coronaviruses and the potential for large-scale clinical applications

In the search for treatment schemes of COVID-19, we start by examining the general weakness of coronaviruses and then identify approved drugs attacking that weakness. The approach, if successful, should identify drugs with a specific mechanism that is at least as effective as the best drugs proposed and are ready for clinical trials. All coronaviruses translate their non-structural proteins ([~]16) in concatenation, resulting in a very large super-protein. Homo-harringtonine (HHT), which has been approved for the treatment of leukemia, blocks protein elongation very effectively. Hence, HHT can repress the replication of many coronaviruses at the nano-molar concentration. In two mouse models, HHT clears SARS-CoV-2 in 3 days, especially by nasal dripping of 40 ug per day. We also use dogs to confirm the safety of HHT delivered by nebulization. The nebulization scheme could be ready for large-scale applications at the onset of the next epidemics. For the current COVID-19, a clinical trial has been approved by the Ditan hospital of Beijing but could not be implemented for want of patients. The protocol is available to qualified medical facilities.

microbiology↗

bioRxiv2020

Reduced function of the glutathione S-transferase S1 suppresses behavioral hyperexcitability in Drosophila expressing a mutant voltage-gated sodium channel

Voltage-gated sodium (Nav) channels play a central role in the generation and propagation of action potentials in excitable cells such as neurons and muscles. To determine how the phenotypes of Nav-channel mutants are affected by other genes, we performed a forward genetic screen for dominant modifiers of the seizure-prone, gain-of-function Drosophila melanogaster Nav-channel mutant, paraShu. Our analyses using chromosome deficiencies, gene-specific RNA interference, and single-gene mutants revealed that a null allele of glutathione S-transferase S1 (GstS1) dominantly suppresses paraShu phenotypes. Reduced GstS1 function also suppressed phenotypes of other seizure-prone Nav-channel mutants, paraGEFS+ and parabss. Notably, paraShu mutants expressed 50% less GstS1 than wild-type flies, further supporting the notion that paraShu and GstS1 interact functionally. Introduction of a loss-of-function GstS1 mutation into a paraShu background led to up- and down-regulation of various genes, with those encoding cytochrome P450 (CYP) enzymes most significantly over-represented in this group. Because GstS1 is a fly ortholog of mammalian hematopoietic prostaglandin D synthase, and in mammals CYPs are involved in the oxygenation of polyunsaturated fatty acids including prostaglandins, our results raise the intriguing possibility that bioactive lipids play a role in GstS1-mediated suppression of paraShu phenotypes.

genetics↗