Biology subjects

Zheng, M.

Publications and source records attributed to Zheng, M..

107 records · Page 6Linked to original sources

bioRxiv2020

Evolution of the Bicoid homeodomain required amino acid substitutions in three subdomains and epistatic interactions between them

Changes in regulatory networks generate materials for evolution to create phenotypic diversity. For transcription networks, multiple studies have shown that alterations in binding sites of cis-regulatory elements correlate well with the gain or loss of specific features of the body plan. Less is known about alterations in the amino acid sequences of the transcription factors (TFs) that bind these elements. Here we study the evolution of Bicoid (Bcd), a homeodomain (HD) protein that is critical for anterior embryo patterning in Drosophila. The ancestor of Bcd (AncBcd) emerged after a duplication of a Zerknullt (Zen)-like ancestral protein (AncZB) in a suborder of flies. AncBcd diverged from AncZB, gaining novel transcriptional and translational activities. We focus on the evolution of the HD of AncBcd, which binds DNA and RNA, and is comprised of four subdomains: an N-terminal arm (NT) and three helices; H1, H2, and Recognition Helix (RH). Using chimeras of subdomains and gene rescue assays in Drosophila, we show that robust patterning activity of the Bcd HD (high frequency rescue to adulthood) is achieved only when amino acid substitutions in three separate subdomains (NT, H1, and RH) are combined. Other combinations of subdomains also yield full rescue, but with lower penetrance, suggesting alternative suboptimal activities. Our results suggest a multi-step pathway for the evolution of the Bcd HD that involved intermediate HD sequences with suboptimal activities, which constrained and enabled further evolutionary changes. They also demonstrate critical epistatic forces that contribute to the robust function of a DNA-binding domain.

evolutionary biology↗

bioRxiv2020

Longitudinal Saliva Omics Responses to Immune Perturbation: A Case Study

Saliva omics, a rapidly developing field for non-invasive diagnostics, may be utilized for monitoring very young or elderly populations, as well as individuals in remote locations. In this study, multiple saliva omics from an individual were monitored over 100 timepoints, over three periods involving: (i) hourly sampling over 24 hours without intervention, (ii) hourly sampling over 24 hours including immune system activation using the standard 23-valent pneumococcal polysaccharide vaccine, (iii) daily sampling for 33 days profiling the post-vaccination response. At each timepoint total saliva transcriptome and proteome were profiled, and salivary extracellular vesicles were derived, from which small-RNA sequencing was used to determine RNA, miRNA, piRNA and bacterial RNA components. The two 24-hour periods were used in a paired analysis to reveal vaccination responses. Temporal trends were classified and collective behavior revealed broad immune-responses captured in saliva, both at the innate as well as the adaptive response time frames.

genomics↗

bioRxiv2020

Broad and differential animal ACE2 receptor usage by SARS-CoV-2

The COVID-19 pandemic has caused an unprecedented global public health and economy crisis. The origin and emergence of its causal agent, SARS-CoV-2, in the human population remains mysterious, although bat and pangolin were proposed to be the natural reservoirs. Strikingly, comparing to the SARS-CoV-2-like CoVs identified in bats and pangolins, SARS-CoV-2 harbors a polybasic furin cleavage site in its spike (S) glycoprotein. SARS-CoV-2 uses human ACE2 as its receptor to infect cells. Receptor recognition by the S protein is the major determinant of host range, tissue tropism, and pathogenesis of coronaviruses. In an effort to search for the potential intermediate or amplifying animal hosts of SARS-CoV-2, we examined receptor activity of ACE2 from 14 mammal species and found that ACE2 from multiple species can support the infectious entry of lentiviral particles pseudotyped with the wild-type or furin cleavage site deficient S protein of SARS-CoV-2. ACE2 of human/rhesus monkey and rat/mouse exhibited the highest and lowest receptor activity, respectively. Among the remaining species, ACE2 from rabbit and pangolin strongly bound to the S1 subunit of SARS-CoV-2 S protein and efficiently supported the pseudotyped virus infection. These findings have important implications for understanding potential natural reservoirs, zoonotic transmission, human-to-animal transmission, and use of animal models. ImportanceSARS-CoV-2 uses human ACE2 as primary receptor for host cell entry. Viral entry mediated by the interaction of ACE2 with spike protein largely determines host range and is the major constraint to interspecies transmission. We examined the receptor activity of 14 ACE2 orthologues and found that wild type and mutant SARS-CoV-2 lacking the furin cleavage site in S protein could utilize ACE2 from a broad range of animal species to enter host cells. These results have important implications in the natural hosts, interspecies transmission, animal models and molecular basis of receptor binding for SARS-CoV-2.

microbiology↗

bioRxiv2020

LY6E Restricts the Entry of Human Coronaviruses, including the currently pandemic SARS-CoV-2

C3A is a sub-clone of human hepatoblastoma HepG2 cell line with the strong contact inhibition of growth. We fortuitously found that C3A was more susceptible to human coronavirus HCoV-OC43 infection than HepG2, which was attributed to the increased efficiency of virus entry into C3A cells. In an effort to search for the host cellular protein(s) mediating the differential susceptibility of the two cell lines to HCoV-OC43 infection, we found that ADAP2, GILT and LY6E, three cellular proteins with known activity of interfering virus entry, expressed at significantly higher levels in HepG2 cells. Functional analyses revealed that ectopic expression of LY6E, but not GILT or ADAP2, in HEK 293 cells inhibited the entry of HCoV-OC43. While overexpression of LY6E in C3A and A549 cells efficiently inhibited the infection of HCoV-OC43, knockdown of LY6E expression in HepG2 significantly increased its susceptibility to HCoV-OC43 infection. Moreover, we found that LY6E also efficiently restricted the entry mediated by the envelope spike proteins of other human coronaviruses, including the currently pandemic SARS-CoV-2. Interestingly, overexpression of serine protease TMPRSS2 or amphotericin treatment significantly neutralized the IFITM3 restriction of human coronavirus entry, but did not compromise the effect of LY6E on the entry of human coronaviruses. The work reported herein thus demonstrates that LY6E is a critical antiviral immune effector that controls CoV infection and pathogenesis via a distinct mechanism. ImportanceVirus entry into host cells is one of the key determinants of host range and cell tropism and is subjected to the control by host innate and adaptive immune responses. In the last decade, several interferon inducible cellular proteins, including IFITMs, GILT, ADAP2, 25CH and LY6E, had been identified to modulate the infectious entry of a variety of viruses. Particularly, LY6E was recently identified as host factors to facilitate the entry of several human pathogenic viruses, including human immunodeficiency virus, influenza A virus and yellow fever virus. Identification of LY6E as a potent restriction factor of coronaviruses expands the biological function of LY6E and sheds new light on the immunopathogenesis of human coronavirus infection.

microbiology↗

bioRxiv2020

Computational target fishing by mining transcriptional data using a novel Siamese spectral-based graph convolutional network

Computational target fishing aims to investigate the mechanism of action or the side effects of bioactive small molecules. Unfortunately, conventional ligand-based computational methods only explore a confined chemical space, and structure-based methods are limited by the availability of crystal structures. Moreover, these methods cannot describe cellular context-dependent effects and are thus not useful for exploring the targets of drugs in specific cells. To address these challenges, we propose a novel Siamese spectral-based graph convolutional network (SSGCN) model for inferring the protein targets of chemical compounds from gene transcriptional profiles. Although the gene signature of a compound perturbation only provides indirect clues of the interacting targets, the SSGCN model was successfully trained to learn from known compound-target pairs by uncovering the hidden correlations between compound perturbation profiles and gene knockdown profiles. Using a benchmark set, the model achieved impressive target inference results compared with previous methods such as Connectivity Map and ProTINA. More importantly, the powerful generalization ability of the model observed with the external LINCS phase II dataset suggests that the model is an efficient target fishing or repositioning tool for bioactive compounds.

bioinformatics↗

bioRxiv2020

AutoOmics: An AutoML Tool for Multi-Omics Research

Deep learning is very promising in solving problems in omics research, such as genomics, epigenomics, proteomics, and metabolics. The design of neural network architecture is very important in modeling omics data against different scientific problems. Residual fully-connected neural network (RFCN) was proposed to provide better neural network architectures for modeling omics data. The next challenge for omics research is how to integrate informations from different omics data using deep learning, so that information from different molecular system levels could be combined to predict the target. In this paper, we present a novel multimodal approach that could efficiently integrate information from different omics data and achieve better accuracy than previous approaches. We evaluate our method in four different tasks: drug repositioning, target gene prediction, breast cancer subtyping and cancer type prediction, and all the four tasks achieved state of art performances. The multimodal approach is implemented in AutoGenome V2 and is also powered with all the previous AutoML convenience to facilitate biomedical researchers.

bioinformatics↗

bioRxiv2020

Facing small and biased data dilemma in drug discovery with federated learning

Artificial intelligence (AI) models usually require large amounts of high-quality training data, which is in striking contrast to the situation of small and biased data faced by current drug discovery pipelines. The concept of federated learning has been proposed to utilize distributed data from different sources without leaking sensitive information of these data. This emerging decentralized machine learning paradigm is expected to dramatically improve the success of AI-powered drug discovery. We here simulate the federated learning process with 7 aqueous solubility datasets from different sources, among which there are overlapping molecules with high or low biases in the recorded values. Beyond the benefit of gaining more data, we also demonstrate federated training has a regularization effect making it superior than centralized training on the pooled datasets with high biases. Further, two more cases are studied to test the usability of federated learning in drug discovery. Our work demonstrates the application of federated learning in predicting drug related properties, but also highlights its promising role in addressing the small data and biased data dilemma in drug discovery.

pharmacology and toxicology↗

bioRxiv2020

Automated design and optimization of multitarget schizophrenia drug candidates by deep learning

Complex neuropsychiatric diseases such as schizophrenia require drugs that can target multiple G protein-coupled receptors (GPCRs) to modulate complex neuropsychiatric functions. Here, we report an automated system comprising a deep recurrent neural network (RNN) and a multitask deep neural network (MTDNN) to design and optimize multitargeted antipsychotic drugs. The system successfully generates novel molecule structures with desired multiple target activities, among which high-ranking compound 3 was synthesized, and demonstrated potent activities against dopamine D2, serotonin 5-HT1A and 5-HT2A receptors. Hit expansion based on the MTDNN was performed, 6 analogs of compound 3 were evaluated experimentally, among which compound 8 not only exhibited specific polypharmacology profiles but also showed antipsychotic effect in animal models with low potential for sedation and catalepsy, highlighting their suitability for further preclinical studies. The approach can be an efficient tool for designing lead compounds with multitarget profiles to achieve the desired efficacy in the treatment of complex neuropsychiatric diseases. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=87 SRC="FIGDIR/small/999615v1_ufig1.gif" ALT="Figure 1"> View larger version (27K): org.highwire.dtl.DTLVardef@17046d8org.highwire.dtl.DTLVardef@3b4c2aorg.highwire.dtl.DTLVardef@10590e9org.highwire.dtl.DTLVardef@6cc609_HPS_FORMAT_FIGEXP M_FIG C_FIG

bioinformatics↗

bioRxiv2020

Visibility Graph Based Community Detection for Biological Time Series

MotivationTemporal behavior is an essential aspect of all biological systems. Time series have been previously represented as networks. Such representations must address two fundamental problems: (i) How to create the appropriate network to reflect the characteristics of biological time series. (ii) How to detect characteristic temporal patterns or events as network communities. General methods to detect communities have used metrics to compare the connectivity within a community to the connectivity one would expect in a random model, or assumed a known number of communities, or are based on the betweenness centrality of edges or nodes. However, such methods were not specifically designed for network representations of time series. We introduce a visibility-graph-based method to build networks from different kinds of biological time series and detect temporal communities within these networks. ResultsTo characterize the uneven sampling of typical experimentally obtained biological time series, and simultaneously capture events associated to peaks and troughs, we introduce the Weighted Dual-Perspective Visibility Graph (WDPVG) for time series. To detect communities, we first find the shortest path of the network between start and end nodes to identify nodes which have high intensities. This identifies the main stem of our community detection algorithm. Then, we aggregate nodes outside the shortest path to the nodes found on the main stem based on the closest path length. Through simulation, we demonstrate the validity of our method in detecting community structures on various networks derived from simulated time series. We also confirm its effectiveness in revealing temporal communities in experimental biological time series. Our results suggest our method of visibility graph based community detection can be effective in detecting temporal biological patterns. AvailabilityThe methods of building WDPVG and visibility graph based community detection are available as a module of the open source Python package PyIOmica (https://doi.org/10.5281/zenodo.3691912) with documentation at https://pyiomica.readthedocs.io/en/latest/. The dataset and codes we used in this manuscript are publicly available at https://doi.org/10.5281/zenodo.3693984. Contactgmias@msu.edu

bioinformatics↗

bioRxiv2020

Endonuclease activity of RecJ from extremely alkaliphilic Bacillus alcalophilus

At present, all documented RecJs are exonucleases, degrading single-stranded nucleic acids. Here, we report a novel RecJ, from the extremely alkaliphilic bacterium Bacillus alcalophilus (BaRecJ), which possesses endonuclease activity and can cleave supercoiled DNA. BaRecJ contains the typical DHH and DHHA1 domains, which are conserved in all RecJs, and a functionally unknown PIWI-like domain at the C-terminus. The endonuclease activity originates from the C-terminal domain of BaRecJ which contains PIWI-like domain, and the exonuclease activity from the DHH and DHHA1 domains. Mutational analysis reveals that several important residues affect the endonuclease activity of BaRecJ. Moreover, BaRecJ cleaves specific target sequences at moderate temperature when directed by a phosphorothioate-modified single-stranded DNA (S-modified ssDNA) guide. These findings suggest that BaRecJ is substantially different from any reported RecJs and has the potential to be developed as a new gene editing tool.

microbiology↗

bioRxiv2019

SPOP Promotes Ubiquitination and Degradation of MyD88 to Suppress the Innate Immune Response

As a canonical adaptor for Toll-like receptor (TLR) family, MyD88 has crucial roles in host defence against infection of microbial pathogens and its dysregulation might induce autoimmune diseases. Here we demonstrate that the Cullin 3-based ubiquitin ligase adaptor SPOP recognizes the intermediate domain and degrades chMyD88 through the proteasome pathway. Knockdown or genetic ablation of chSPOP leads to aberrant elevation of the chMyD88 protein. Consequently, ChSPOP negatively regulates the activity of NF-{kappa}B pathway and thus the production of IL-1{beta} and IL-8 upon LPS challenge. Furthermore, SPOP deficiency mice are more susceptible to infection of Salmonella typhimurium. Collectively, these findings demonstrate chMyD88 as a bona fide substrate of chSPOP and uncover a mechanism by which chSPOP suppresses the innate immune signaling.\n\nAuthor SummaryMyD88 is a central adaptor mediating the initiate of innate immune response and production of proinflammatory cytokines that restrain pathogens and activate adaptive immunity. Although MyD88 is crucial for the host to prevent pathogenic infection, misregulation of MyD88 abundance might lead to autoimmune diseases. Thus, degradation of MyD88 is a canonical mechanism to terminate cytokines production. Here we characterized a novel E3 ligase SPOP that target MyD88 for degradation. SPOP attenuated IL1{beta} and IL8 production through K48-linked polyubiquitination and degradation of MyD88 and thus impaired immune responses. SPOP deficient mice show more susceptibility to infection by Salmonella typhimurium. These findings demonstrate that SPOP is a negative regulator of MyD88-dependent pathway activation triggered by LPS and Salmonella typhimurium, which helps the host to maintain immune homeostasis.

immunology↗

bioRxiv2019

Expression Profile of Circular RNAs in Epicardial Adipose Tissue in Heart Failure

Recent studies have reported circular RNA (circRNA) expression profiles in various tissue types; specifically, a recent work showed a detailed circRNA expression landscape in the heart. However, circRNA expression profile in human epicardial adipose tissue (EAT) remains undefined. RNA-sequencing was carried out to compare circRNA expression patterns in EAT specimens from coronary artery disease (CAD) cases between the heart failure (HF) and non-HF groups. The top highly expressed EAT circRNAs corresponded to genes involved in cell proliferation and inflammatory response, including KIAA0182, RHOBTB3, HIPK3, UBXN7, PCMTD1, N4BP2L2, CFLAR, EPB41L2, FCHO2, FNDC3B and SPECC1. Among the 141 circRNAs substantially different between the HF and non-HF groups (P<0.05;fold change>2), hsa_circ_0005565 stood out, and was mostly associated with positive regulation of metabolic processes and insulin resistancein GO and KEGG pathway analyses, respectively. These data indicate EAT circRNAs contribute to the pathogenesis of metabolic disorders causing HF.

bioinformatics↗

bioRxiv2019

Multiplex chromatin interaction analysis by signal processing and statistical algorithms

The single-molecule multiplex chromatin interaction data generated by emerging non-ligation-based 3D genome mapping technologies provide novel insights into high dimensional chromatin organization, yet introduce new computational challenges. We developed MIA-Sig (https://github.com/TheJacksonLaboratory/mia-sig.git), an algorithmic framework to de-noise the data, assess the statistical significance of chromatin complexes, and identify topological domains and inter-domain contacts. On chromatin immunoprecipitation (ChIP)-enriched data, MIA-Sig can clearly distinguish the protein-associated interactions from the non-specific topological domains.

genomics↗

bioRxiv2019

Basement membrane regulates fibronectin organization using sliding focal adhesions driven by a contractile winch

We have discovered that basement membrane and its major components can induce rapid, strikingly robust fibronectin organization. In this new matrix assembly mechanism, 5{beta}1 integrin-based focal adhesions slide actively on the underlying matrix towards the ventral cell center through the dynamic shortening of myosin IIA-associated actin stress fibers to drive rapid fibronectin fibrillogenesis distal to the adhesion. This mechanism contrasts with classical fibronectin assembly based on stable/fixed-position focal adhesions containing V{beta}3 integrins plus 5{beta}1 integrin translocation into proximal fibrillar adhesions. On basement membrane components, these sliding focal adhesions contain standard focal adhesion constituents but completely lack classical V{beta}3 integrins. Instead, peripheral 3{beta}1 or 2{beta}1 adhesions mediate initial cell attachment, but over time are switched to 5{beta}1 integrin-based sliding focal adhesions to assemble fibronectin matrix. This basement membrane-triggered mechanism produces rapid fibronectin fibrillogenesis, providing a mechanistic explanation for the well-known widespread accumulation of fibronectin at many organ basement membranes.

cell biology↗

bioRxiv2019

ChIA-DropBox: a novel analysis and visualization pipeline for multiplex chromatin interactions

ChIA-Drop is a new experimental method for mapping multiplex chromatin interactions with single-molecule precision by barcoding chromatin complexes inside microfluidics droplets, followed by pooled DNA sequencing. The chromatin DNA reads with the same droplet-specific barcodes are inferred to be derived from the same chromatin interaction complex. Here, we describe an integrated computational pipeline, named ChIA-DropBox, that is specifically designed for reconstructing chromatin reads in each droplet and refining multiplex chromatin complexes from raw ChIA-Drop sequencing reads, and then visualizing the results. First, ChIA-DropBox maps and filters sequencing reads, and then reconstructs the \"chromatin droplets\" by parsing the barcode sequences and grouping together chromatin reads with the same barcode. Based on the concept of chromosome territories that most chromatin interactions take place within the same chromosome, potential mixing up of chromatin complexes derived from different chromosomes could be readily identified and separated. Accordingly, ChIA-DropBox refines these \"chromatin droplets\" into purely intra-chromosomal chromatin complexes, ready for downstream analysis. For visualization, ChIA-DropBox converts the ChIA-Drop data to pairwise format and automatically generates input files for viewing 2D contact maps in Juicebox and viewing loops in BASIC Browser. Finally, ChIA-DropBox introduces a new browser, named ChIA-View, for interactive visualization of multiplex chromatin interactions.

bioinformatics↗

bioRxiv2018

Abundance and community composition of comammox bacteria in different ecosystems by a universal primer set

Complete ammonia oxidizing bacteria (CAOB) have been recognized as a new member of ammonia-oxidizing microorganisms (AOMs) due to its single-step nitrification capability. However, the abundance and diversity of CAOB in environmental ecosystems were still far from known owing to the lack of specific molecular marker. Herein, a universal primer set specifically targeting the both clades of CAOB amoA gene with high specificity and coverage was successfully designed. Intriguingly, real-time quantitative PCR tests revealed that CAOB were ubiquitous and unexpectedly abundant in agricultural soils, river sediments, intertidal zones, drinking water and wastewater treatment systems. Phylogenetic analysis indicated that clade A existed in all the five types of ecosystems, while clade B were only detected in soil and sediment samples. Four sub-clusters were further classified within clade A, in which N. nitrosa cluster dominated CAOB amoA in activated sludge samples while the new recognized soil cluster was the primary constitute in soils. Moreover, the niche specialization between different CAOB species and the environmental conditions were supposed to be the primary driven force to shape the diversity and community of CAOB. This study provided a strong evidence in support of the ubiquities and high abundances of CAOB in various environmental ecosystems and highlighted the significance of including CAOB as the new member of AOMs to re-evaluate the biogeochemical nitrogen cycle.

microbiology↗

bioRxiv2018

Multiplex Chromatin Interaction Analysis with Single-Molecule Precision

We describe a microfluidics-based strategy for genome-wide analysis of multiplex chromatin interactions with single-molecule precision. In multiplex chromatin interaction analysis (multi-ChIA), individual chromatin complexes are partitioned into droplets that contain a gel bead with unique DNA barcode, in which tethered chromatin DNA fragments are barcoded and amplified for sequencing and mapping to demarcate chromatin contacts. Thus, multi-ChIA has the unprecedented ability to uncover multiplex chromatin interactions at single-molecule level, which has been impossible using previous methods that rely on analyzing pairwise contacts via proximity ligation. We demonstrate that multiplex chromatin interactions predominantly contribute to topologically associated domains, and clusters of gene promoters and enhancers provide a fundamental topological framework for co-transcriptional regulation.

genomics↗