Biology subjects

Wei, L.

Publications and source records attributed to Wei, L..

At least 73 records · Page 4Linked to original sources

bioRxiv2024

Loop Grafting Improves Neuraminidase Protein Vaccine Production

Influenza virus neuraminidase is a crucial target for protective antibodies, yet the development of recombinant neuraminidase protein as a vaccine has been held back by instability and variable expression. We have taken a pragmatic approach to improving expression and stability of neuraminidase by grafting antigenic surface loops from low-expressing neuraminidase proteins onto the scaffold of high-expressing counterparts. The resulting hybrid proteins retained the antigenic properties of the loop donor while benefiting from the high-yield expression, stability, and tetrameric structure of the loop recipient. These hybrid proteins were recognised by a broad set of human monoclonal antibodies elicited by influenza infection or vaccination, with X-ray structures validating the accurate structural conformation of the grafted loops and the enzymatic cavity. Immunisation of mice with neuraminidase hybrids induced inhibitory antibodies to the loop donor and protected against lethal influenza challenge. This pragmatic technique offers a robust solution for improving the expression and stability of influenza neuraminidase proteins for vaccine development. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=200 SRC="FIGDIR/small/617814v3_ufig1.gif" ALT="Figure 1"> View larger version (57K): org.highwire.dtl.DTLVardef@14d4b4eorg.highwire.dtl.DTLVardef@18af09aorg.highwire.dtl.DTLVardef@ba1982org.highwire.dtl.DTLVardef@1e0d143_HPS_FORMAT_FIGEXP M_FIG C_FIG

immunology↗

bioRxiv2024

Evaluation of scalp-based targeting methods of DLPFC for TMS therapy: EEG cap F3, Beam F3, CPC F3, and adjusted Beam F3

BackgroundRepetitive transcranial magnetic stimulation targeting the left dorsolateral prefrontal cortex (DLPFC) offers a promising approach for treating depression. Scalp-based targeting methods, including EEG Cap, Beam F3, CPC F3, and adjusted Beam F3, have evolved to improve clinical application. ObejectivesTo evaluate and compare the accuracy, reliability, speed, and training efficacy of these four methods. MethodsThree trained technicians conducted repeated DLPFC targeting measurements on 10 3D-printed head models. Accuracy was assessed by calculating targeting error and electric field strength at the subgenual anterior cingulate cortex anti-correlated peak. Reliability was evaluated through inter- and intra-rater variability, and training efficacy was examined by comparing targeting errors between trained technicians and novices. ResultsCPC F3, Beam F3, and adjusted Beam F3 demonstrated comparable accuracy in terms of electric field strength, but CPC F3 had the lowest targeting error (4.2 mm). CPC F3 also exhibited the highest reliability (inter-rater variability at 3.3 mm, intra-rater variability at 2.3 mm). The EEG cap was the fastest method (80 seconds), while CPC F3 was faster than Beam F3 but slower than EEG Cap. No significant differences in targeting error between trained technicians and novices were noted with CPC F3 and adjusted Beam F3. ConclusionThe comparative analysis of four scalp-based targeting methods provides insights that can enhance clinical decision-making and the practical application. While accuracy among CPC F3, Beam F3, and adjusted Beam F3 was comparable, CPC F3 emerges as highly practical due to its significantly better reliability, training efficacy, moderate speed, and straightforward implementation.

neuroscience↗

bioRxiv2024

Dynamic epigenomic landscape of carbon-concentrating mechanisms in the model industrial oleaginous microalga Nannochloropsis oceanica

Despite their ecological and physiological significance, how carbon-concentrating mechanisms (CCM) are regulated in microalgae remains elusive. Here in the model industrial microalga Nannochloropsis oceanica, we uncovered an epigenetic regulatory mechanism for CCM via comprehensive, multi-dimensional epigenomic analyses. Our integrated study reveals the complex interplay among histone modifications, dynamic nucleosome positioning, and 3D chromatin structure in regulating gene expression during low CO2 adaptation, despite minimal DNA methylation. Histone modifications, including lysine acetylation (H3K9ac and H3K27ac), crotonylation (Kcr), and methylation (H3K4me2), were associated with active chromatin states. Significantly altered ChIP-Seq peaks were linked to 43.1% of the differentially expressed genes (DEGs). Notably, H3K4me2 exhibited a distinct dual-peak profile around the transcription start site (TSS), which is unique among microalgae and plants. Chromatin compartment dynamics were correlated with gene expression and histone modifications, particularly H3K4me2, while differentially positioned nucleosomes were associated with key CCM-related genes and transcription factors. To further elucidate the role of H3K4me2, we knocked out its methyltransferase, resulting in genome-wide H3K4me2 peak shifts, slower growth, and reduced photosynthesis. These changes were accompanied by differential expression of key genes of NoHINT and NoPMA2, whose subsequent deletion and overexpression revealed their subtle yet significant impacts on growth and photosynthetic efficiency under low CO2 conditions, with NoHINT regulating growth and NoPMA2 influencing photosynthesis. Finally, we proposed a comprehensive model for epigenetic regulation of CCM in N. oceanica, which established a foundation for enhancing microalgal productivity through targeted epigenetic modifications. HighlightsO_LIMulti-layered epigenetic modifications contribute to the regulation of CCM and its adaptation to low CO2 in N. oceanica. C_LIO_LIThe histone modification H3K4me2 regulates the growth and photosynthesis of N. oceanica under low CO2 conditions. C_LIO_LIH3K4me2 targets NoHINT and NoPMA2 in N. oceanica, with NoHINT influencing growth dynamics and NoPMA2 modulating photosynthetic efficiency under low CO2 conditions. C_LI

genomics↗

bioRxiv2024

Crystal structure and catalytic mechanism of PL35 family glycosaminoglycan lyases with an ultrabroad substrate spectrum

Recently, a new class of glycosaminoglycan (GAG) lyases (GAGases) belonging to PL35 family has been discovered with an ultrabroad substrate spectrum that can degrade three types of uronic acid-containing GAGs (hyaluronic acid, chondroitin sulfate and heparan sulfate) or even alginate. In this study, the structures of GAGase II from Spirosoma fluviale and GAGase VII from Bacteroides intestinalis DSM 17393 were determined at 1.9 and 2.4 [A] resolution, respectively, and their catalytic mechanism was investigated by the site-directed mutant of their crucial residues and molecular docking assay. Structural analysis showed that GAGase II and GAGase VII consist of an N-terminal (/)6 toroid multidomain and a C-terminal two-layered {beta}-sheet domain with Mn2+. Notably, although GAGases share similar folds and catalytic mechanisms with some GAG lyases and alginate lyases, they exhibit higher structural similarity with alginate lyases than GAG lyases, which may present a crucial structural evidence for the speculation that GAG lyases with (/)n toroid and antiparallel {beta}-sheet structures arrived by a divergent evolution from alginate lyases with the same folds. Overall, this study not only solved the structure of PL35 GAG lyases for the first time and investigated their catalytic mechanism, especially the reason why GAGase III can additionally degrade alginate, but also provided a key clue in the divergent evolution of GAG lyases that originated from alginate lyases.

biochemistry↗

bioRxiv2024

Estrus-Tracking Cortical Neurons Integrate Social cues and Reproductive states to Adaptively Control Sexually Dimorphic Sociosexual Behaviors

Female sociosexual behaviors, essential for survival and reproduction, are adaptively modulated by ovarian hormones. However, the neural mechanisms integrating internal hormonal states with external social cues to guide these behaviors remain poorly understood. Here we identified primary estrous-sensitive Cacna1h-expressing medial prefrontal (mPFCCacna1h+) neurons that orchestrate adaptive sociosexual behaviors. Bidirectional manipulation of mPFCCacna1h+ neurons drives opposite-sex-directed behavioral shifts between estrus and diestrus females. In males, these neurons serve opposite functions compared to estrus females, mediating sexually dimorphic effects via anterior hypothalamic outputs. Miniscope imaging reveals mixed-representation of self-estrous states and social target sex in distinct mPFCCacna1h+ subpopulations, with biased-encoding of opposite-sex social cues in estrus females and males. Mechanistically, ovarian hormone-driven upregulation of Cacna1h-encoded T-type calcium channels underlies estrus-specific activity changes and sexual-dimorphic function of mPFCCacna1h+ neurons. These findings uncover a prefrontal circuit that integrates internal hormonal states and target-sex information to exert sexually bivalent top-down control over adaptive social behaviors.

neuroscience↗

bioRxiv2024

Regulation of angiogenesis by signal sequence-derived peptides

BackgroundThe neuropilin-like, Discoidin, CUB and LCCL domain containing 2 (DCBLD2) is a transmembrane protein with an unusually long signal sequence (SS) composed of N-terminal (N) and C-terminal (C) subdomains, separated by a transition (tra) subdomain. DCBLD2 interacts with VEGFR-2 and regulates VEGF-induced endothelial cell signaling, proliferation and migration, as well as angiogenesis. The exact mechanisms by which DCBLD2 interacts with VEGFR2 to modulate VEGF signaling remain unclear. MethodsSearching for VEGFR2 interacting DCBLD2 domains, we generated various constructs containing different DCBLD2 domain combinations and conducted co-immunoprecipitation and signaling studies in HEK 293T and endothelial cells. Several peptides were synthesized based on the identified domain, and their effect on VEGF signaling was assessed in vitro in cell culture and in vivo using matrigel plug and corneal micropocket assays. The effect of the lead peptide was further evaluated using a murine hindlimb ischemia model. ResultsDCBLD2 SS interacted with VEGFR2 and promoted VEGF signaling. SS was not cleaved in the mature DCBLD2 and its hydrophobic transmembrane traC segment, but not the N subdomain, was involved in DCBLD2-VEGFR2 interaction. The smallest unit in DCBLD2 SS that interacts with VEGFR2 was the L5VL5 sequence. Even after the central valine was removed, the L10 sequence mimicked the DCBLD2 SS traCs effect on VEGF-signaling, while shorter or longer poly-leucine sequences were less effective. Finally, a synthetic traC peptide enhanced VEGF signaling in vitro, promoted VEGF-induced angiogenesis in vivo, and improved blood flow recovery following hindlimb ischemia. ConclusionDCBLD2 SS along with its derivative peptides can promote VEGFR2 signaling and angiogenesis. Synthetic peptides based on DCBLD2 SS hold promise as therapeutic agents for regulating angiogenesis. Importantly these findings refine the traditional view of signal sequences as mere targeting elements, revealing a role in cellular signaling. This opens new avenues for research and therapeutic strategies.

cell biology↗

bioRxiv2024

Exploring the diversity of anti-defense systems across prokaryotes, phages, and mobile genetic elements

The co-evolution of prokaryotes, phages, and mobile genetic elements (MGEs) over the past billions of years has driven the emergence and diversification of defense and anti-defense systems alike. Anti-defense proteins have diverse functional domains, sequences, and are typically small, creating a challenge to detect anti-defense homologs across the prokaryotic genomes. To date, no tools comprehensively annotate anti-defense proteins within a desired genome or MGE. Here, we developed "AntiDefenseFinder" - a free open-source tool and web service that detects 156 anti-defense systems (of one or more proteins) in any genomic sequence. Using this dataset, we identified 47,981 anti-defense systems distributed across prokaryotes, phage, and MGEs. We found that some genes co-localize in "anti-defense islands", including E. coli T4 and Lambda phages, although many are standalone. Out of the 112 systems detected in bacteria, 100 systems localize only or preferentially in prophages, plasmids, phage satellites, integrons, and integrative and conjugative elements. However, over 80% of anti-Pycsar protein 1 (Apyc1) resides in non-mobile regions of bacteria. Evolutionary and functional analyses revealed that Apyc1 likely originated in bacteria to regulate cNMP signaling, but was co-opted multiple times by phages to overcome cNMP-utilizing defenses. With the AntiDefenseFinder tool, we hope to facilitate the identification of the full repertoire of anti-defense systems in MGEs, the discovery of new protein functions, and a deeper understanding of host-pathogen arms race.

microbiology↗

bioRxiv2024

Systems-level reconstruction of kinase phosphosignaling networks regulating endothelial barrier integrity using temporal data

Phosphosignaling networks control cellular processes. We built kinase-mediated regulatory networks elicited by thrombin stimulation of brain endothelial cells using two computational strategies: Temporal Pathway Synthesizer (TPS), which uses phosphoproteomics data as input, and Temporally REsolved KInase Network Generation (TREKING), which uses kinase inhibitor screens. TPS and TREKING predicted overlapping barrier-regulatory kinases connected with unique network topology. Each strategy effectively describes regulatory signaling networks and is broadly applicable across biological systems.

systems biology↗

bioRxiv2024

Wide-field bond-selective fluorescence imaging: from single-molecule to cellular imaging beyond video-rate

Wide-field (WF) imaging is pivotal for observing dynamic biological events. While WF chemical microscopy offers high molecular specificity, it lacks the sensitivity for single-molecule detection. In contrast, WF fluorescence microscopy provides live-cell dynamic mapping but fails to leverage the rich chemical information necessary for functional interpretations. To address these limitations, we introduce Wide-Field Bond-selective Fluorescence-detected Infrared-Excited (WF-BonFIRE) spectro-microscopy. This technique combines rationally optimized imaging speed and field-of-view (FOV) to achieve single-molecule sensitivity with bond-selective contrast. WF-BonFIRE outperforms its point-scanning counterpart, enhancing frame acquisition up to 10,000 times. We demonstrate WF-BonFIREs capabilities in imaging cells, astrocytes, and live neurons, capturing single FOVs up to 50 {micro}m x 50 {micro}m, with further expansion via multi-FOV mosaicking. Additionally, we have implemented a temporal-delay modulation scheme that allows real-time kilohertz imaging speeds up to 1500 Hz. This enables millisecond temporal resolution while monitoring random motion of live Escherichia coli. Overall, WF-BonFIRE significantly broadens the possibilities for chemical imaging, enabling high-speed observations at unparalleled sensitivity levels. One-Sentence SummaryWide-field bond-selective fluorescence imaging pushes chemical-sensitive microscopy platform into a new regime, achieving single-molecule sensitivity and speeds up to kilohertz.

biophysics↗

bioRxiv2024

Slow and steady: auditory features for discriminating animal vocalizations

We propose that listeners can use temporal regularities - spectro-temporal correlations that change smoothly over time - to discriminate animal vocalizations within and between species. To test this idea, we used Slow Feature Analysis (SFA) to find the most temporally regular components of vocalizations from birds (blue jay, house finch, American yellow warbler, and great blue heron), humans (English speakers), and rhesus macaques. We projected vocalizations into the learned feature space and tested intra-class (same speaker/species) and inter-class (different speakers/species) auditory discrimination by a trained classifier. We found that: 1) Vocalization discrimination was excellent (> 95%) in all cases; 2) Performance depended primarily on the [~]10 most temporally regular features; 3) Most vocalizations are dominated by [~]10 features with high temporal regularity; and 4) These regular features are highly correlated with the most predictable components of animal sounds.

neuroscience↗

bioRxiv2024

Transcranial Brain Atlas Based on Photon Measurement Density Function in a Triple-Parameter Standard Channel Space

Functional near-infrared spectroscopy (fNIRS) is a widely used transcranial brain imaging technique in neuroscience research. Nevertheless, its lack of anatomical information poses challenges for designing appropriate optode montage and localizing fNIRS signals to underlying anatomical regions. The photon measurement density function (PMDF) is often employed to address these issues, as it accurately measures the sensitivity of a fNIRS signal channel to perturbations of absorption coefficients at any brain locations. However, existing PMDF-based localization methods have two limitations: a limited channel space, and PMDF estimation based on single standard head models, which differ anatomically from individual subjects. To overcome these limitations, this study proposes a continuous fNIRS standard channel space and constructs a PMDF-based transcranial brain atlas (PMDF-TBA) by calculating PMDFs based on MRI images of 48 adults. PMDF-TBA contains group-averaged sensitivities of channels to gray matter and brain regions of atlases, such as Brodmann, AAL2, and LPBA40. Through leave-one-out cross-validation, we evaluated the prediction ability of PMDF-TBA for sensitivity of unseen individuals and found that it outperformed PMDFs based on single standard head models. Thus, PMDF-TBA serves as a more generalizable fNIRS spatial localization tool. Therefore, PMDF-TBA can be utilized to optimize optode montage design, improve channel sensitivity to target brain regions, and assist in the source localization of fNIRS data, thereby promoting the application of fNIRS in neuroscience research.

neuroscience↗

bioRxiv2024

Mapping Endothelial-Macrophage Interactions in Diabetic Vasculature: Role of TREM2 in Vascular Inflammation and Ischemic Response

Diabetes mellitus (DM) significantly accelerates vascular diseases like peripheral arterial disease (PAD). Endothelial cells (ECs) and macrophages (M{Phi}s) singularly and synergistically are important contributors to DM-associated vascular dysfunction. Single-cell (sc) profiling technologies are revealing the true heterogeneity of ECs and M{Phi}s, but how this cellular diversity translates to cell-cell interactions, and consequentially vascular function, remains unknown. We leveraged scRNA sequencing and spatial transcriptome (ST) profiling to analyze human mesenteric arteries from non-diabetic (ND) and type 2 diabetic (T2D) donors. We generated a transcriptome and interactome map encompassing the major arterial cells and highlighted Triggering Receptor Expressed on Myeloid Cells 2 (TREM2) as a top T2D-induced gene in mononuclear phagocytes (MPs), with concomitant increases of TREM2 ligands in ECs. We verified DM-associated TREM2 induction in cell and mouse models, and found that TREM2 inhibition decreases pro-inflammatory responses in MPs and ECs, as well as increases EC migration in vitro. Furthermore, TREM2 inhibition using a neutralizing antibody enhanced ischemic recovery and flow reperfusion in DM mice subjected to hindlimb ischemia, suggesting that TREM2 promotes ischemic injury in DM. Finally, in human PAD, co-existing DM was associated with greater expression of TREM2 and its interaction with ECs, with a further increase in ischemic tissue compared to patient-matched non-ischemic tissue. Collectively, our study presents the first atlas of human diabetic vessels with single cell and spatial resolution, and identifies TREM2-EC interaction as a key driver of diabetic vasculopathies, the targeting of which may offer an opportunity to ameliorate vascular dysfunction associated with DM-PAD.

cell biology↗

bioRxiv2024

Using machine learning to dissect host kinases required for Leishmania internalization and development

The Leishmania life cycle alternates between promastigotes, found in the sandfly, and amastigotes, found in mammals. When an infected sandfly bites a host, promastigotes are engulfed by phagocytes (i.e., neutrophils, dendritic cells, and macrophages) to establish infection. When these phagocytes die or break down, amastigotes must be re-internalized to survive within the acidic phagolysosome and establish disease. To define host kinase regulators of Leishmania promastigote and amastigote uptake and survival within macrophages, we performed an image-based kinase regression screen using a panel of 38 kinase inhibitors with unique yet overlapping kinase targets. We also targeted inert beads to complement receptor 3 (CR3) or Fc{gamma} receptors (FcR) as controls by coating them with complement/C3bi or IgG respectively. Through this approach, we identified several putative host kinases that regulate receptor-mediated phagocytosis and/or the uptake of L. amazonensis. Findings included kinases previously implicated in Leishmania uptake (such as Src family kinases (SFK), Abl family kinases (ABL1/c-Abl, ABL2/Arg), and spleen tyrosine kinase (SYK)), but we also uncovered many novel kinases. Our methods also predicted host kinases necessary for promastigotes to convert to amastigotes or for amastigotes to survive within macrophages. Overall, our results suggest that the concerted action of multiple interconnected networks of host kinases are needed over the course of Leishmania infection, and that the kinases required for the parasites life cycle may differ substantially depending on which receptors are bound and the life cycle stage that is internalized. In addition, using our screen, we identified kinases that appear to preferentially regulate the uptake of parasites over beads, indicating that the methods required for Leishmania to be internalized by macrophages may differ significantly from generalized phagocytic mechanisms. Our findings are intended to be used as a hypothesis generation resource for the broader scientific community studying the roles of kinases in host-pathogen interactions.

microbiology↗

bioRxiv2024

DenoiseST: A dual-channel unsupervised deep learning-based denoising method to identify spatial domains and functionally variable genes in spatial transcriptomics

Spatial transcriptomics provides a unique opportunity for understanding cellular organization and function in a spatial context. However, spatial transcriptome exists the problem of dropout noise, exposing a major challenge for accurate downstream data analysis. Here, we proposed DenoiseST, a dual-channel unsupervised adaptive deep learning-based denoising method for data imputing, clustering, and identifying functionally variable genes in spatial transcriptomics. To leverage spatial information and gene expression profiles, we proposed a dual-channel joint learning strategy with graph convolutional networks to sufficiently explore both linear and nonlinear representation embeddings in an unsupervised manner, enhancing the discriminative information learning ability from the global perspectives of data distributions. In particular, DenoiseST enables the adaptively fitting of different gene distributions to the clustered domains and employs tissue-level spatial information to accurately identify functionally variable genes with different spatial resolutions, revealing their enrichment in corresponding gene pathways. Extensive validations on a total of 18 real spatial transcriptome datasets show that DenoiseST obtains excellent performance and results on brain tissue datasets indicate it outperforms the state-of-the-art methods when handling artificial dropout noise with a remarkable margin of [~]15%, demonstrating its effectiveness and robustness. Case study results demonstrate that when applied to identify biological structural regions on human breast cancer spatial transcriptomic datasets, DenoiseST successfully detected biologically significant immune-related structural regions, which are subsequently validated through Gene Ontology (GO), cell-cell communication, and survival analysis. In conclusion, we expect that DenoiseST is a novel and efficient method for spatial transcriptome analysis, offering unique insights into spatial organization and function.

bioinformatics↗

bioRxiv2024

Discovering and overcoming the bias in neoantigen identification by unified machine learning models

Neoantigens play a crucial role in tumor immune process and precisely identifying them can greatly contribute to tumor immunotherapy design. There are three main steps in the neoantigen immune process, i.e., binding with MHCs, extracellular presentation, and immunogenicity induction. Various computational methods have been developed, but the overall accuracy of neoantigen identification remains relatively low. Here, we established a unified transformer-based framework ImmuBPI that comprised three tasks. Cross-task model interpretation discovered a counterfactual pattern learned by the immunogenicity prediction model overlooked in previous studies. We demonstrated that this model bias arose from training data imbalance and the neglect of bias control would not only mislead the models but also hinder existing benchmarks from fair evaluation. We further designed a mutual information-based debiasing strategy that showed preliminary potential in alleviating the bias. We believe these observations will provide insightful perspectives for future neoantigen prediction by high-lighting the necessity of bias control.

bioinformatics↗

bioRxiv2024

scMulan: a multitask generative pre-trained language model for single-cell analysis

Cells can be viewed as complex stories written by coordinated expression of genes. The success of AI large language models (LLMs) in mastering the human language inspired us to develop a large AI model scMulan with 368 million parameters to generate cell transcriptomics with designated attributes by learning the cell language. We defined a unified c-sentence to incorporate cell transcriptomics and meta-attributes, and pre-trained scMulan on the equivalence of 100 million human cells. Experiments showed that scMulan can generate designated pseudo transcriptomics, predict missing attributes of cells, reconstruct unobserved cells along functional gradients, and can help to identify driving regulators of cell fates. The generated data passed tests of current tools and can reflect the underlying biology.

bioinformatics↗

bioRxiv2024

PPTC7 limits mitophagy through proximal and dynamic interactions with BNIP3 and NIX

PPTC7 is a mitochondrial-localized PP2C phosphatase that maintains mitochondrial protein content and metabolic homeostasis. We previously demonstrated that knockout of Pptc7 elevates mitophagy in a BNIP3- and NIX-dependent manner, but the mechanisms by which PPTC7 influences receptor-mediated mitophagy remain ill-defined. Here, we demonstrate that loss of PPTC7 upregulates BNIP3 and NIX post-transcriptionally and independent of HIF-1 stabilization. On a molecular level, loss of PPTC7 prolongs the half-life of BNIP3 and NIX while blunting their accumulation in response to proteasomal inhibition, suggesting that PPTC7 promotes the ubiquitin-mediated turnover of BNIP3 and NIX. Consistently, overexpression of PPTC7 limits the accumulation of BNIP3 and NIX protein levels in response to pseudohypoxia, a well-known inducer of mitophagy. This PPTC7-mediated suppression of BNIP3 and NIX protein expression requires an intact PP2C catalytic motif but is surprisingly independent of its mitochondrial targeting, indicating that PPTC7 influences mitophagy outside of the mitochondrial matrix. We find that PPTC7 exists in at least two distinct states in cells: a longer isoform, which likely represents full length protein, and a shorter isoform, which likely represents an imported, matrix-localized phosphatase pool. Importantly, anchoring PPTC7 to the outer mitochondrial membrane is sufficient to blunt BNIP3 and NIX accumulation, and proximity labeling and fluorescence co-localization experiments suggest that PPTC7 associates with BNIP3 and NIX within the native cellular environment. Importantly, these associations are enhanced in cellular conditions that promote BNIP3 and NIX turnover, demonstrating that PPTC7 is dynamically recruited to BNIP3 and NIX to facilitate their degradation. Collectively, these data reveal that a fraction of PPTC7 dynamically localizes to the outer mitochondrial membrane to promote the proteasomal turnover of BNIP3 and NIX.

biochemistry↗

bioRxiv2024

scDirect: key transcription factor identification for directing cell state transitions based on single-cell multi-omics data

Cell state transitions are complicated processes that occur in various life activities. Understanding and artificially manipulating them have been longstanding challenges. Substantial experiments reveal that the transitions could be directed by several key transcription factors (TFs). Here we present scDirect, a computational framework to identify key TFs based on single-cell RNA-seq and ATAC-seq data. scDirect models the TF identification task as a linear inverse problem, and solve it with gene regulatory networks enhanced by a graph attention network. Through a benchmarking on a single-cell human embryonic stem cell atlas, we demonstrate the robustness and superiority of scDirect against alternative analysis methods on TF identification. We apply scDirect on various datasets, and scDirect exhibits high capability in identifying key TFs in cell differentiation and somatic cell conversion. Furthermore, scDirect can efficiently identify TF combinations for cell reprogramming, many of which have been experimentally validated. We envision that scDirect can utilize rapidly increasing single-cell datasets to identify key TFs for directing cell state transitions and may become an effective tool to facilitate cell engineering and regenerative medicine.

bioinformatics↗