Biology subjects

Wu, H.

Publications and source records attributed to Wu, H..

At least 379 records · Page 21Linked to original sources

bioRxiv2022

ACh signaling modulates activity of the GABAergic signaling network in the basolateral amygdala and behavior in stress-relevant paradigms.

Balance between excitatory and inhibitory (E/I) signaling is important for maintaining homeostatic function in the brain. Indeed, dysregulation of inhibitory GABA interneurons in the amygdala has been implicated in human mood disorders. We hypothesized that acetylcholine (ACh) signaling in the basolateral amygdala (BLA) might alter E/I balance resulting in changes in stress-sensitive behaviors. We therefore measured ACh release as well as activity of calmodulin-dependent protein kinase II (CAMKII)-, parvalbumin (PV)-, somatostatin (SOM)- and vasoactive intestinal protein (VIP)-expressing neurons in the BLA of awake, behaving mice. ACh levels and activity of both excitatory and inhibitory BLA neurons increased when animals were actively coping, and decreased during passive coping, in the light-dark box, tail suspension and social defeat. Changes in neuronal activity preceded behavioral state transitions, suggesting that BLA activity may drive the shift in coping strategy. In contrast to exposure to escapable stressors, prolonging ACh signaling with a cholinesterase antagonist changed the balance of activity among BLA cell types, significantly increasing activity of VIP neurons and decreasing activity of SOM cells, with little effect on CaMKII or PV neurons. Knockdown of 7 or {beta}2-containing nAChR subtypes in PV and SOM, but not CaMKII or VIP, BLA neurons altered behavioral responses to stressors, suggesting that ACh signaling through nAChRs on GABA neuron subtypes contributes to stress-induced changes in behavior. These studies show that ACh modulates the GABAergic signaling network in the BLA, shifting the balance between SOM, PV, VIP and CaMKII neurons, which are normally activated coordinately during active coping in response to stress. Thus, prolonging ACh signaling, as occurs in response to chronic stress, may contribute to maladaptive behaviors by shifting the balance of inhibitory signaling in the BLA.

neuroscience↗

bioRxiv2022

Inflammasome mediated neuronal-microglial crosstalk: a therapeutic substrate in C9orf72-FTD/ALS.

Intronic G4C2 hexanucleotide repeat expansions (HRE) of C9orf72 are the most common cause of familial variants of frontotemporal dementia/amyotrophic lateral sclerosis (FTD/ALS). G4C2 HREs in C9orf72 undergo non-canonical repeat-associated translation, producing dipeptide repeat (DPR) proteins, with various deleterious impacts on cellular homeostasis. While five different DPRs are produced, poly(glycine-arginine) (GR) is amongst the most toxic and is the only DPR to accumulate in the associated clinically relevant anatomical locations of the brain. Previous work has demonstrated the profound effects of a poly(GR) model of C9orf72 FTD/ALS, including motor impairment, memory deficits, neurodegeneration, and neuroinflammation. Neuroinflammation is hypothesized to be a driving factor in the disease course; microglia activation is present prior to symptom onset and persists throughout the disease. Here, using an established mouse model of C9orf72 FTD/ALS we investigate the contributions of the nod-like receptor pyrin-containing 3 (NLRP3) inflammasome in the pathogenesis of FTD/ALS. We find that inflammasome-mediated neuroinflammation is increased with microglial activation, cleavage of caspase-1, production of IL-1{beta} and upregulation of Cxcl10 in the brain of C9orf72 FTD/ALS mice. Excitingly, we find that genetic ablation of Nlrp3 significantly improved survival, protected behavioral deficits and prevented neurodegeneration suggesting a novel mechanism involving HRE-mediated induction of innate immunity. The findings provide experimental evidence of the integral role of HRE in inflammasome-mediated innate immunity in the C9orf72 variant of FTD/ALS pathogenesis and suggest the NLRP3 inflammasome as a therapeutic target.

neuroscience↗

bioRxiv2022

Multimodal single cell sequencing of human diabetic kidney disease implicates chromatin accessibility and genetic background in disease progression

Multimodal single cell sequencing is a powerful tool for interrogating cell-specific changes in transcription and chromatin accessibility. We performed single nucleus RNA (snRNA-seq) and assay for transposase accessible chromatin sequencing (snATAC-seq) on human kidney cortex from donors with and without diabetic kidney disease (DKD) to identify altered signaling pathways and transcription factors associated with DKD. Both snRNA-seq and snATAC-seq had an increased proportion of VCAM1+ injured proximal tubule cells (PT_VCAM1) in DKD samples. PT_VCAM1 has a pro-inflammatory expression signature and transcription factor motif enrichment implicated NFkB signaling. We used stratified linkage disequilibrium score regression to partition heritability of kidney-function-related traits using publicly-available GWAS summary statistics. Cell-specific PT_VCAM1 peaks were enriched for heritability of chronic kidney disease (CKD), suggesting that genetic background may regulate chromatin accessibility and DKD progression. snATAC-seq found cell-specific differentially accessible regions (DAR) throughout the nephron that change accessibility in DKD and these regions were enriched for glucocorticoid receptor (GR) motifs. Changes in chromatin accessibility were associated with decreased expression of insulin receptor, increased gluconeogenesis, and decreased expression of the GR cytosolic chaperone, FKBP5, in the diabetic proximal tubule. Cleavage under targets and release using nuclease (CUT&RUN) profiling of GR binding in bulk kidney cortex and an in vitro model of the proximal tubule (RPTEC) showed that DAR co-localize with GR binding sites. CRISPRi silencing of GR response elements (GRE) in the FKBP5 gene body reduced FKBP5 expression in RPTEC, suggesting that reduced FKBP5 chromatin accessibility in DKD may alter cellular response to GR. We developed an open-source tool for single cell allele specific analysis (SALSA) to model the effect of genetic background on gene expression. Heterozygous germline single nucleotide variants (SNV) in proximal tubule ATAC peaks were associated with allele-specific chromatin accessibility and differential expression of target genes within cis-coaccessibility networks. Partitioned heritability of proximal tubule ATAC peaks with a predicted allele-specific effect was enriched for eGFR, suggesting that genetic background may modify DKD progression in a cell-specific manner.

genomics↗

bioRxiv2022

An essential function for autocrine Hedgehog signaling in epithelial proliferation and differentiation in the trachea

The tracheal epithelium is a primary target for pulmonary diseases as it provides a conduit for air flow between the environment and the lung lobes. The cellular and molecular mechanisms underlying airway epithelial cell proliferation and differentiation remain poorly understood. Hedgehog (Hh) signaling orchestrates communication between epithelial and mesenchymal cells in the lung, where it modulates stromal cell proliferation, differentiation and signaling back to the epithelium. Here, we reveal a new, autocrine function of Hh signaling in airway epithelial cells. Epithelial cell depletion of the ligand Sonic hedgehog (SHH) or its effector Smoothened (SMO) causes defects in both epithelial cell proliferation and differentiation. In cultured primary human airway epithelial cells, Hh signaling inhibition also hampers cell proliferation and differentiation. Epithelial Hh function is mediated, at least in part, through transcriptional activation as Hh signaling inhibition leads to downregulation of cell-type specific transcription factor genes in both the mouse trachea and human airway epithelial cells. These results provide new insights into the role of Hh signaling in epithelial cell proliferation and differentiation during airway development.

developmental biology↗

bioRxiv2022

Teasing apart the joint effect of demography and natural selection in the birth of a contact zone

Vast population movements induced by recurrent climatic cycles have shaped the genetic structure of plant species. This is especially true in Scandinavia that was repeatedly glaciated. During glacial periods trees were confined to refugia, south and east of the ice sheet, from which they recolonized Scandinavia as the ice melted away. This multi-pronged recolonization led to large contact zones in most species. We leverage large genomic data from 5000 trees to reconstruct the demographic history of Norway spruce (Picea abies) and test for the presence of natural selection during the recolonization process and the establishment of the contact zone. Sweden is today made up of two large genetic clusters, a southern one originating from the Baltics and a Northern one originating from Northern Russia. The contact zone delineating these two clusters closely matches the limit between two major climatic regions. This suggests that natural selection contributed to the establishment and the maintenance of the contact zone. To test this hypothesis we first used Approximate Bayesian Computation; an Isolation-with migration model with genome-wide linked selection fits the data better than a purely neutral one. Secondly, we identified loci characterized by both extreme allele frequency differences between geographic regions and association to the variables defining the climatic zones. These loci, many of which are related to phenology, form clusters present on all linkage groups. Altogether, the current genetic structure reflects the joint effect of climatic cycles, recolonization and selection on the establishment of strong local adaptation and contact zones. Significance StatementUnderstanding how past climatic events, human actions and evolutionary forces contributed to the present distribution of genetic diversity is crucial to predict their reaction to the current climate crisis. Vast distribution shifts induced by past environmental changes, local ecological processes, natural selection and human transfers contributed to the current distribution of Norway spruce across Northern Europe. Genome-wide polymorphisms from thousands of individuals show that Scandinavia was recolonized after the Last Glacial from both south and north. This two-pronged recolonization established a contact zone between two genetic clusters that matches the limit between two major climate zones. The contact zone is shaped and maintained by natural selection on a large number of loci that form blocks of co-adapted loci spread genome-wide.

evolutionary biology↗

bioRxiv2022

Novel modification of Luminex assay for characterization of extracellular vesicle populations in biofluids.

Most approaches to extracellular vesicle (EV) characterization focus on EV size or density. However, such approaches provide few clues regarding EV origin, molecular composition, and function. New methods to characterize the EV surface proteins may aid our understanding of their origin, physiological roles, and biomarker potential. Recently developed immunoassays for intact EVs based on ELISA, NanoView, SIMOA and MesoScale platforms are highly sensitive, but have limited multiplexing capabilities, whereas MACSPlex FACS enables the detection of multiple EV surface proteins, but requires significant quantities of purified EVs, which limits its adoption. Here, we describe a novel Luminex-based immunoassay, which combines multiplexing capabilities with high sensitivity and, importantly, bypasses the enrichment and purification steps that require larger sample volumes. We demonstrate the methods specificity for detecting EV surface proteins using multiple EV depletion techniques, EVs of specific cellular origin isolated from culture media, and by co-localization with established EV surface markers. Using this novel approach, we elucidate differences in the tetraspanin profiles of the EVs carrying erythrocyte and neuron markers. Using size exclusion chromatography, we show that plasma EVs of putative neuronal and tissue macrophage origin are eluted in fractions distinct from those derived from erythrocytes, or from their respective cultured cells. In conclusion, our novel multiplexed assay differentiates between EVs from erythrocytes, macrophages, and neurons, and offers a new means for capture, classification, and profiling of EVs from diverse sources.

cell biology↗

bioRxiv2022

Oxytocin modulates social brain network correlations in resting and task state

The effects of oxytocin (OT) on the social brain can be tracked upon assessing the neural activity in resting and task states, and developing a system-level framework for characterizing the state-based functional relationships of its distinct effect. Here, we contribute to this framework by examining how OT modulates social brain network correlations during resting and task states, using fMRI. First, we investigated network activation, followed by an analysis of the relationships between networks and individual differences. Subsequently, we evaluated the functional connectivity in both states. Finally, the relationship between networks across states was represented by the predictive power of networks in the resting state for task-evoked activities. The differences in the predicted accuracy between the subjects displayed individual variations in this relationship. Our results showed that the activity of the dorsal default mode network (DDMN) in the resting state had the largest predictive power for task-evoked activation of the precuneus network (PN) only in the OT group. The results also demonstrated that OT reduced the individual variation in PN in the prediction process. These findings suggest a distributed but modulatory effect of OT on the association between resting and task-dependent brain networks.

neuroscience↗

bioRxiv2022

NeoRdRp: A comprehensive dataset for identifying RNA-dependent RNA polymerase of various RNA viruses from metatranscriptomic data

RNA viruses are distributed throughout various environments, and most RNA viruses have recently been identified by metatranscriptome sequencing. However, due to the high nucleotide diversity of RNA viruses, it is still challenging to identify novel RNA viruses from metatranscriptome data. To overcome this issue, we created a dataset of RNA-dependent RNA polymerase (RdRp) domains that are essential for all RNA viruses belonging to Orthornavirae. Genes with RdRp domains from various RNA viruses were clustered based on their amino acid sequence similarity. For each cluster, a multiple sequence alignment was generated, and a hidden Markov model (HMM) profile was created if the number of sequences was greater than three. We further refined the 426 HMM profiles by detecting the RefSeq RNA virus sequences and subsequently combined the hit sequences with the RdRp domains. As a result, a total of 1,182 HMM profiles were generated from 12,502 RdRp domain sequences, and the dataset was named NeoRdRp. Almost all NeoRdRp HMM profiles successfully detected RdRp domains, specifically in the UniProt dataset. Furthermore, we compared the NeoRdRp dataset with two previously reported methods for RNA virus detection using metatranscriptome sequencing data. Our methods successfully identified most of the RNA viruses in the datasets; however, some RNA viruses were not detected, as in the cases of the other two methods. The NeoRdRp can be repeatedly improved by adding new RdRp sequences and is applicable as a system for detecting various RNA viruses from diverse metatranscriptome data.

microbiology↗

bioRxiv2021

Fusion protein strategies for cryo-EM study of G protein-coupled receptors

Single particle cryogenic-electron microscopy (cryo-EM) is used extensively to determine structures of activated G protein-coupled receptors (GPCRs) in complex with G proteins or arrestins. However, applying it to GPCRs without signaling proteins remains challenging because most receptors lack structural features in their soluble domains to facilitate image alignment. In GPCR crystallography, inserting a fusion protein between transmembrane helices 5 and 6 is a highly successful strategy for crystallization. Although the similar strategy has the potential to broadly facilitate cryo-EM structure determination of GPCRs alone without signaling protein, the critical determinants that make this approach successful are not yet clear. Here, we address this shortcoming by exploring different fusion protein designs, which led to structures of antagonist bound A2A adenosine receptor at 3.4[A] resolution and unliganded Smoothened at 3.7[A] resolution. The fusion strategies explored here are likely applicable to cryo-EM interrogation of other GPCRs and small integral membrane proteins.

biophysics↗

bioRxiv2021

PCBP1 Deficient Pigs Hold the Potential to Inhibit CSFV Infection

Classical swine fever virus (CSFV), pathogen of classic swine fever, has caused severe economic losses worldwide. Poly (rC)-binding protein 1 (PCBP1), interacting with Npro of CSFV, plays a vital role in CSFV growth. Here, our research is the first report to generate PCBP1 knockout pigs via gene editing technology. The PCBP1 knockout pigs exhibited normal birth weight, reproductive-performance traits, and developed normally. Viral challenge results indicated that primary cells isolated from F0 and F1 generation pigs could significantly reduce CSFV infection. Additional mechanism exploration further confirmed that PCBP1 KO mediated antiviral effect is related with the activation of type I interferon. Beyond showing that gene editing strategy can be used to generate PCBP1 KO pigs, our study introduces a valuable animal model for further investigating infection mechanisms of CSFV that help to develop better antiviral solution. ImportanceAs a negative regulator in immune modulation, the effects of PCBP1 on viral replication have been found to be valuable. Here, this study was the first report to generate PCBP1 knockout pigs with normal pregnancy rate and viability. Primary cells isolated from F0 and F1 generation PCBP1 knockout pigs could significantly reduce CSFV infection. The PCBP1 knockout pigs could be used as a natural host models for investigating the effects of PCBP1-mediating critical interactions on viral replication and helping to develop better antiviral solution.

molecular biology↗

bioRxiv2021

Nucleocapsid 203 mutations enhance SARS-CoV-2 immune evasion

Previous work indicated that the nucleocapsid 203 mutation increase the virulence and transmission of the SARS-CoV-2 Alpha variant. However, Delta later outcompeted Alpha and other lineages, promoting a new wave of infections. Delta also possesses a nucleocapsid 203 mutation, R203M. Large-scale epidemiological analyses suggest a synergistic effect of the 203 mutation and the spike L452R mutation, associated with Delta expansion. Viral competition experiments demonstrate the synergistic effect in fitness and infectivity. More importantly, we found that the combination of R203M and L452R brings in a 3.2-fold decrease in neutralizing titers to the neutralizing serum relative to L452R-only virus. R203M/L452R show an increased fitness after the initiation of global vaccination programmes, possibly associated with the enhanced immune evasion. Another rapidly emerging variant Omicron also bears the 203 mutation. Thus, we proposed that nucleocapsid mutations play an essential role for the rise and predominance of variants in concern.

microbiology↗

bioRxiv2021

A discovery-based proteomics approach identifies protein disulfide isomerase (PDIA1) as a biomarker of β cell stress in type 1 diabetes

BackgroundActivation of stress pathways intrinsic to the {beta} cell are thought to both accelerate {beta} cell death and increase {beta} cell immunogenicity in type 1 diabetes (T1D). However, information on the timing and scope of these responses is lacking. MethodsTo identify temporal and disease-related changes in islet {beta} cell protein expression, data independent acquisition-mass spectrometry was performed on islets collected longitudinally from NOD mice and NOD-SCID mice rendered diabetic through T cell adoptive transfer. FindingsIn islets collected from female NOD mice at 10, 12, and 14 weeks of age, we found a time-restricted upregulation of proteins involved in the maintenance of {beta} cell function and stress mitigation, followed by loss of expression of protective proteins that heralded diabetes onset. Pathway analysis identified EIF2 signaling and the unfolded protein response, mTOR signaling, mitochondrial function, and oxidative phosphorylation as commonly modulated pathways in both diabetic NOD mice and NOD-SCID mice rendered acutely diabetic by adoptive transfer, highlighting this core set of pathways in T1D pathogenesis. In immunofluorescence validation studies, {beta} cell expression of protein disulfide isomerase A1 (PDIA1) and 14-3-3b were found to be increased during disease progression in NOD islets, while PDIA1 plasma levels were increased in pre-diabetic NOD mice and in the serum of children with recent-onset T1D compared to age and sex-matched non-diabetic controls. InterpretationWe identified a common and core set of modulated pathways across distinct mouse models of T1D and identified PDIA1 as a potential human biomarker of {beta} cell stress in T1D.

physiology↗

bioRxiv2021

Lessons on fruiting body morphogenesis from genomes and transcriptomes of Agaricomycetes

Fruiting bodies of mushroom-forming fungi (Agaricomycetes) are among the most complex structures produced by fungi. Unlike vegetative hyphae, fruiting bodies grow determinately and follow a genetically encoded developmental program that orchestrates tissue differentiation, growth and sexual sporulation. In spite of more than a century of research, our understanding of the molecular details of fruiting body morphogenesis is limited and a general synthesis on the genetics of this complex process is lacking. In this paper, we aim to comprehensively identify conserved genes related to fruiting body morphogenesis and distill novel functional hypotheses for functionally poorly characterized genes. As a result of this analysis, we report 921 conserved developmentally expressed gene families, only a few dozens of which have previously been reported in fruiting body development. Based on literature data, conserved expression patterns and functional annotations, we provide informed hypotheses on the potential role of these gene families in fruiting body development, yielding the most complete description of molecular processes in fruiting body morphogenesis to date. We discuss genes related to the initiation of fruiting, differentiation, growth, cell surface and cell wall, defense, transcriptional regulation as well as signal transduction. Based on these data we derive a general model of fruiting body development, which includes an early, proliferative phase that is mostly concerned with laying out the mushroom body plan (via cell division and differentiation), and a second phase of growth via cell expansion as well as meiotic events and sporulation. Altogether, our discussions cover 1480 genes of Coprinopsis cinerea, and their orthologs in Agaricus bisporus, Cyclocybe aegerita, Armillaria ostoyae, Auriculariopsis ampla, Laccaria bicolor, Lentinula edodes, Lentinus tigrinus, Mycena kentingensis, Phanerochaete chrysosporium, Pleurotus ostreatus, and Schizophyllum commune, providing functional hypotheses for [~]10% of genes in the genomes of these species. Although experimental evidence for the role of these genes will need to be established in the future, our data provide a roadmap for guiding functional analyses of fruiting related genes in the Agaricomycetes. We anticipate that the gene compendium presented here, combined with developments in functional genomics approaches will contribute to uncovering the genetic bases of one of the most spectacular multicellular developmental processes in fungi.

evolutionary biology↗

bioRxiv2021

A fungal plant pathogen overcomes conserved broad-spectrum disease resistance by rapid gene loss

Hosts and pathogens typically engage in a co-evolutionary arms race. This also applies to phytopathogenic powdery mildew fungi, which can rapidly overcome plant resistance and perform host jumps. Using experimental evolution, we show that the powdery mildew pathogen Blumeria hordei is capable of breaking the agriculturally important broad-spectrum resistance conditioned by barley loss-of-function mlo mutants. Partial mlo virulence of evolved B. hordei isolates is correlated with a distinctive pattern of adaptive mutations, including small-sized (8-40 kb) deletions, of which one is linked to the de novo insertion of a transposable element. Occurrence of the mutations is associated with a transcriptional induction of effector proteinencoding genes that is absent in mlo-avirulent isolates on mlo mutant plants. The detected mutational spectrum comprises the same loci in at least two independently isolated mlo-virulent isolates, indicating convergent multigenic evolution. The mutational events emerged in part early (within the first five asexual generations) during experimental evolution, likely generating a founder population in which incipient mlo virulence was later stabilized by additional events. This work highlights the rapid dynamic genome evolution of an obligate biotrophic plant pathogen with a transposon-enriched genome.

plant biology↗

bioRxiv2021

Unconventional secretion of unglycosylated ORF8 is critical for the cytokine storm during SARS-CoV-2 infection

Coronavirus disease 2019 is a respiratory infectious disease caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Evidence on the pathogenesis of SARS-CoV-2 is accumulating rapidly. In addition to structural proteins such as Spike and Envelope, the functional roles of non-structural and accessory proteins in regulating viral life cycle and host immune responses remain to be understood. Here, we show that open reading frame 8 (ORF8) acts as messenger for inter-cellular communication between alveolar epithelial cells and macrophages during SARS-CoV-2 infection. Mechanistically, ORF8 is a secretory protein that can be secreted by infected epithelial cells via both conventional and unconventional secretory pathways. The unconventionally secreted ORF8 recognizes the IL17RA receptor of macrophages and induces cytokine release. However, conventionally secreted ORF8 cannot bind to IL17RA due to N-linked glycosylation. Furthermore, we found that Yip1 interacting factor homolog B (YIF1B) is a channel protein that translocates unglycosylated ORF8 into vesicles for unconventional secretion. Blocking the unconventional secretion of ORF8 via a YIF1B knockout in hACE2 mice attenuates inflammation and yields delayed mortality following SARS-CoV-2 challenge.

microbiology↗

bioRxiv2021

Lateral plate mesoderm cell-based organoid system for NK cell regeneration from human pluripotent stem cells

Human pluripotent stem cell (hPSC)-induced NK (iNK) cells are a promising "off-the-shelf" cell product for universal immune therapy. Conventional methods for iNK cell regeneration from hPSCs include embryonic body-formation and feeder-based expansion steps, which bring instability, time-consuming, and high costs for manufacture. In this study, we develop an embryonic body-free, organoid aggregate method for NK cell regeneration from hPSCs. In a short time window of 27-day induction, millions of hPSC input can produce over billions of iNK cells without the necessity of NK cell-expansion feeders. The iNK cells highly express classical toxic granule proteins, apoptosis-inducing ligands, as well as abundant activating and inhibitory receptors. Functionally, the iNK cells eradicate human tumor cells by mechanisms of direct cytotoxity, apoptosis, and antibody-dependent cellular cytotoxicity. This study provides a reliable scale-up method for regenerating human NK cells from hPSCs, which promotes the universal availability of NK cell products for immune therapy.

immunology↗

bioRxiv2021

Genome assembly of the JD17 soybean provides a new reference genome for Comparative genomics

Cultivated soybean (Glycine max) is an important source for protein and oil. Many elite cultivars with different traits have been developed for different conditions. Each soybean strain has its own genetic diversity, and the availability of more high-quality soybean genomes can enhance comparative genomic analysis for identifying genetic underpinnings for its unique traits. In this study, we constructed a high-quality de novo assembly of an elite soybean cultivar Jidou 17 (JD17) with chromsome contiguity and high accuracy. We annotated 52,840 gene models and reconstructed 74,054 high-quality full-length transcripts. We performed a genome-wide comparative analysis based on the reference genome of JD17 with three published soybeans (WM82, ZH13 and W05), which identified five large inversions and two large translocations specific to JD17, 20,984 - 46,912 PAVs spanning 13.1 - 46.9 Mb in size, and 5 - 53 large PAV clusters larger than 500kb. 1,695,741 - 3,664,629 SNPs and 446,689 - 800,489 Indels were identified and annotated between JD17 and them. Symbiotic nitrogen fixation (SNF) genes were identified and the effects from these variants were further evaluated. It was found that the coding sequences of 9 nitrogen fixation-related genes were greatly affected. The high-quality genome assembly of JD17 can serve as a valuable reference for soybean functional genomics research.

genomics↗

bioRxiv2021

HybridDTA: Hybrid Data Fusion through Pairwise Training for Drug-Target Affinity Prediction

Candidate compounds with high binding affinities toward a target protein are likely to be developed as drugs. Deep neural networks (DNNs) have attracted increasing attention for drug-target affinity (DTA) estimation owning to their efficiency. However, the negative impact of batch effects caused by measure metrics, system technologies, and other assay information is seldom discussed when training a DNN model for DTA. Suffering from the data deviation caused by batch effects, the DNN models can only be trained on a small amount of "clean" data. Thus, it is challenging for them to provide precise and consistent estimations. We design a batch-sensitive training framework, namely BatchDTA, to train the DNN models. BatchDTA implicitly aligns multiple batches toward the same protein, alleviating the impact of the batch effects on the DNN models. Extensive experiments demonstrate that BatchDTA facilitates four mainstream DNN models to enhance the ability and robustness on multiple DTA datasets. The average concordance index (CI) of the DNN models achieves a relative improvement of 4.0%. BatchDTA can also be applied to the fused data collected from multiple sources to achieve further improvement.

bioinformatics↗