Biology subjects

Cheng, X.

Publications and source records attributed to Cheng, X..

At least 109 records · Page 6Linked to original sources

bioRxiv2024

Novel derivatives of BCV and (S)-HPMPA inhibit orthopoxviruses and human adenoviruses more potently than BCV

BCV and tecovirimat are the only two chemical drugs that have been approved to treat smallpox and can be requested for Mpox treatment through a single-patient Emergency Investigational New Drug (EIND) application. Disappointedly, the efficacy of tecovirimat manifested in a recent clinical trial is far from being satisfactory, while the clinical efficacy of BCV is still inconclusive. Given that MPXV, variola and other emerging orthopoxviruses are posing serious threats to global health, it is urgent to develop better therapeutics. In this study, we tested the antiviral effects of three novel prodrugs, which were designed based on previously reported parent drugs, either (S)-HPMPC (cidofovir) or (S)-HPMPA. We found that one of the (S)-HPMPA-based prodrugs, ODE-(S)-HPMPA formate, exhibited significantly better anti-orthopoxvirus activity than BCV both in vitro and in vivo, which also inhibited human adenovirus type 2 and type 21 more efficiently than BCV. Most strikingly, the EC50 and EC90 of ODE-(S)-HPMPA formate against MPXV were more than 40-fold lower than those of BCV. In contrast, we observed that the anti-HSV-1 activities of the (S)-HPMPA-based prodrugs were less effective than those of the cidofovir-based prodrugs (BCV and BCV formate), especially in vivo. Moreover, we showed for the first time that cytidine and adenine analog combined therapies could provide mice with complete protection against lethal challenges of both vaccinia and HSV-1. Collectively, we propose that both the ODE-(S)-HPMPA formate and the BCV/ODE-(S)-HPMPA formate combination are worth further investigations for their potential clinical applications.

microbiology↗

bioRxiv2024

Multi-Modal Large Language Model Enables Protein Function Prediction

Predicting the functions of proteins can greatly accelerate biological discovery and applications, where deep learning methods have recently shown great potential. However, these methods predominantly predict protein functions as discrete categories, which fails to capture the nuanced and complex nature of protein functions. Furthermore, existing methods require the development of separate models for each prediction task, a process that can be both resource-heavy and time-consuming. Here, we present ProteinChat, a versatile, multi-modal large language model that takes a proteins amino acid sequence as input and generates comprehensive narratives describing its function. ProteinChat is trained using over 1,500,000 (protein, prompt, answer) triplets curated from the Swiss-Prot dataset, covering diverse functions. This novel model can universally predict a wide range of protein functions, all within a single, unified framework. Furthermore, ProteinChat supports interactive dialogues with human users, allowing for iterative refinement of predictions and deeper exploration of protein functions. Our experimental results, evaluated through both human expert assessment and automated metrics, demonstrate that ProteinChat outperforms general-purpose LLMs like GPT-4, one of the flagship LLMs, by over ten-fold. In addition, ProteinChat exceeds or matches the performance of task-specific prediction models.

bioinformatics↗

bioRxiv2024

RIPK4 driven by TP53 mutations promotes resistance to redox stress of CRC by phosphorylating MTHFD1

This study investigates advanced colorectal cancer (CRC), focusing on its tendency for distant metastasis and chemotherapy resistance. It highlights the importance of PANoptosis, a cell death pathway, and the role of the Receptor-interacting serine/threonine-protein kinase (RIPK) family in tumor progression. RIPK4s tissue-specific functions in cancer cell behavior are emphasized, including its influence on invasion, migration, and oxidative stress resistance. The study reveals the critical balance of reactive oxygen species (ROS) in cancer cells, linked to antioxidant defenses and NADPH production for survival. A key finding is the connection between TP53 mutations in CRC and increased RIPK4 expression, which enhances MTHFD1 phosphorylation, boosts NADPH production, reduces ROS, and promotes resistance to PANoptosis, leading to metastasis. The research identifies the molecular basis of CRC metastasis, showing how RIPK4 regulates MTHFD1 to resist PANoptosis, offering new therapeutic targets for metastatic CRC and potential improvements in patient outcomes.

molecular biology↗

bioRxiv2024

C-COMPASS: A Neural Network Tool for Multi-Omic Classification of Cell Compartments

Functional compartmentalization in eukaryotic cells is essential for maintaining physiological processes. The development of systematic organelle proteomic techniques, such as Protein Correlation Profiling (PCP) and Localization of Organelle Proteins by Isotope Tagging (LOPIT), has enhanced our understanding of organelle dynamics and protein localization. However, the complexity of the data and the need for advanced computational skills limit their accessibility. To address this, we introduce C-COMPASS, an open-source software featuring a user-friendly interface that utilizes a neural network-based regression model to predict the spatial distribution of proteins across cellular compartments. C-COMPASS manages complex multilocalization patterns and integrates protein abundance information to model organelle composition changes under various biological conditions. Using C-COMPASS, we mapped the organelle proteomic landscapes of humanized liver mice in different metabolic states and modeled changes in organelle composition to provide insights into cellular adaptations. Additionally, we extended cellular maps to the lipid level by co-generating protein and lipid profiles. C-COMPASS trains neural networks with marker protein profiles to predict lipid localizations, enabling parallel mapping of lipid and protein localization. This approach overcomes the lack of knowledge of organelle-specific lipid markers and identifies previously unknown organelle-specific lipid species. C-COMPASS offers a comprehensive solution for studying organelle dynamics at the multi-omics level, designed to be accessible without requiring extensive computational expertise or specialized high-performance computing equipment.

cell biology↗

bioRxiv2024

Impaired experience-dependent inter-areal network connectivity across the visual cortex in Fmr1 KO mice

Fragile X syndrome (FX) is the most prevalent inheritable form of autism spectrum disorder (ASD), characterized by hypersensitivity, difficulty in habituating to new sensory stimuli, and intellectual disability. Individuals with FX often experience visual perception and learning deficits. Visual experience leads to the emergence of the familiarity-evoked theta band oscillations in the primary visual cortex (V1) and the lateromedial area (LM) of mice. These theta oscillations in V1 and LM are synchronized with each other, providing a mechanism of sensory multi-areal binding. However, how this multi-areal binding and the corresponding theta oscillations are altered in FX is not known. Using iDISCO whole brain clearing with light-sheet microscopy, we quantified immediate early gene Fos expression in V1 and LM, identifying deficits in experience-dependent neural activity in FX mice. We performed simultaneous in vivo recordings with silicon probes in V1 and LM of awake mice and channelrhodopsin-2-assisted circuit mapping (CRACM) in acute brain slices to examine the neural activity and strength of long-range synaptic connections between V1 and LM in both wildtype (WT) and Fmr1 knockout (KO) mice, the model of FX, before and after visual experience. Our findings reveal synchronized familiarity-evoked theta oscillations in V1 and LM, the increased strength of V1[->]LM functional and synaptic connections, which correlated with the corresponding changes of presynaptic short-term plasticity in WT mice. The LM oscillations were attenuated in FX mice and correlated with impaired functional and synaptic connectivity and short-term plasticity in the feedforward (FF) V1[->]LM and feedback (FB) LM[->]V1 pathways. Finally, using 4Pi single-molecule localization microscopy (SMLM) in thick brain tissue, we identified experience-dependent changes in the density and shape of dendritic spines in layer 5 pyramidal cells of WT mice, which correlated with the functional synaptic measurements. Interestingly, there was an increased dendritic spine density and length in naive FX mice that failed to respond to experience. Our study provides the first comprehensive characterization of the role of visual experience in triggering inter-areal neural synchrony and shaping synaptic connectivity in WT and FX mice.

neuroscience↗

bioRxiv2024

Down regulation of Engase in Caenorhabditis elegans may improve its stresses adaptivity

Endo-beta-N-acetylglucosaminidase (ENGASE) is one of the key enzymes involved in the regulation of structure and function of glycoproteins. It is conserved from prokaryotic to eukaryotic cells. Although their activities in vitro and applications have been well studied, the biological function of ENGASE remains to be illustrated. In this study, we analyzed the molecular and physiological function of Engase from Caenorhabditis elegans homolog eng-1(CeEngase). We found that CeEngase knockout or knockdown increased the environmental stresses adaptability, such as heat stress and osmotic stress. Preliminary glycomics analysis showed that the basement membrane proteins of extracellular matrix may be the main targets of CeENGASE. In addition, CeENGASE may selectively prefer to N2H7 glycans on glycoproteins. In conclusion, our data illustrated that the defection and/or down regulation of CeEngase may provide a beneficially adaptation for stresses.

biochemistry↗

bioRxiv2024

A Multitrait Genome-Wide Association Study Reveals a Requirement for the Strigolactone Receptor DWARF14 in Optimal GOLVEN10 Signaling

GOLVEN/ROOT MERISTEM GROWTH FACTOR family of signaling peptides have been shown to control root lateral organ number, density and positioning in plants, although the signaling pathways involved remain obscure. A diverse set of 171 Medicago truncatula HapMap accessions with variation in responses to the GOLVEN 10 peptide, GLV10, were used to identify 74 significant loci controlling seven traits related to nodule formation and root architecture. Importantly, a single nucleotide polymorphism (SNP) in the upstream region of the MtGLV10 peptide-inducible strigolactone receptor gene, MtDWARF14 was significantly associated with insensitivity of nodule density to GLV10, suggesting a link between strigolactone signaling and GLV10 responsiveness. Three independent d14 mutants of the DWARF14 gene were found to hypernodulate, while overexpression of the gene led to reduction in nodule number, phenocopying GLV10. A null mutant, mtd14-1, remained sensitive to GLV10s effect on nodule density. However, at the transcriptional level, the mutant failed to effectively induce the expression of the GOLVEN marker genes, MtPLETHORA3 and MtPINLIKES2. Our study uncovers a hitherto unknown link between the strigolactone and GLV peptide signaling pathways using genotype x environment analysis of Medicago HapMap lines and provides a putative molecular mechanism for recovery from frost damage to fine roots.

plant biology↗

bioRxiv2024

MSAGPT: Neural Prompting Protein Structure Prediction via MSA Generative Pre-Training

Multiple Sequence Alignment (MSA) plays a pivotal role in unveiling the evolutionary trajectories of protein families. The accuracy of protein structure predictions is often compromised for protein sequences that lack sufficient homologous information to construct high-quality MSA. Although various methods have been proposed to generate virtual MSA under these conditions, they fall short in comprehensively capturing the intricate co-evolutionary patterns within MSA or require guidance from external oracle models. Here we introduce MSAGPT, a novel approach to prompt protein structure predictions via MSA generative pre-training in the low-MSA regime. MSAGPT employs a simple yet effective 2D evolutionary positional encoding scheme to model the complex evolutionary patterns. Endowed by this, its flexible 1D MSA decoding framework facilitates zero-or few-shot learning. More-over, we demonstrate that leveraging the feedback from AlphaFold2 can further enhance the models capacity via Rejective Fine-tuning (RFT) and Reinforcement Learning from AF2 Feedback (RLAF). Extensive experiments confirm the efficacy of MSAGPT in generating faithful virtual MSA to enhance the structure prediction accuracy (up to +8.5% TM-Score on few-shot scenarios). The transfer learning capabilities also highlight its great potential for facilitating other protein tasks.

bioinformatics↗

bioRxiv2024

Training Compute-Optimal Protein Language Models

We explore optimally training protein language models, an area of significant interest in biological research where guidance on best practices is limited. Most models are trained with extensive compute resources until performance gains plateau, focusing primarily on increasing model sizes rather than optimizing the efficient compute frontier that balances performance and compute budgets. Our investigation is grounded in a massive dataset consisting of 939 million protein sequences. We trained over 300 models ranging from 3.5 million to 10.7 billion parameters on 5 to 200 billion unique tokens, to investigate the relations between model sizes, training token numbers, and objectives. First, we observed the effect of diminishing returns for the Causal Language Model (CLM) and that of overfitting for the Masked Language Model (MLM) when repeating the commonly used Uniref database. To address this, we included metagenomic protein sequences in the training set to increase the diversity and avoid the plateau or overfitting effects. Second, we obtained the scaling laws of CLM and MLM on Transformer, tailored to the specific characteristics of protein sequence data. Third, we observe a transfer scaling phenomenon from CLM to MLM, further demonstrating the effectiveness of transfer through scaling behaviors based on estimated Effectively Transferred Tokens. Finally, to validate our scaling laws, we compare the large-scale versions of ESM-2 and PROGEN2 on downstream tasks, encompassing evaluations of protein generation as well as structure- and function-related tasks, all within less or equivalent pre-training compute budgets.

bioinformatics↗

bioRxiv2024

Impact of Parental Time-Restricted Feeding on Offspring Metabolic Phenotypic Traits

A substantial body of research elucidates the mechanisms and health advantages associated with intermittent fasting (IF). However, the impact of parental IF on offspring remains unclear. Through an investigation involving four IF and ad libitum combinations of parental mating groups, we demonstrate that parental IF (daily time-restricted feeding) influences offsprings metabolic health indicators in male and female offspring in distinct ways. We found that when both parents are on IF their offspring exhibit protection against the adverse effects of a high-fat, high-sugar, and high-salt diet in a sex-specific manner. This study underscores the potential significance of parental lifestyle modifications involving dietary restriction for the metabolic status of their children and their risk for obesity and diabetes.

genetics↗

bioRxiv2024

Role of the CTCF Binding Site in Human T-Cell Leukemia Virus-1 Pathogenesis

During HTLV-1 infection, the virus integrates into the host cell genome as a provirus with a single CCCTC binding protein (CTCF) binding site (vCTCF-BS), which acts as an insulator between transcriptionally active and inactive regions. Previous studies have shown that the vCTCF-BS is important for maintenance of chromatin structure, regulation of viral expression, and DNA and histone methylation. Here, we show that the vCTCF-BS also regulates viral infection and pathogenesis in vivo in a humanized (Hu) mouse model of adult T-cell leukemia/lymphoma. Three cell lines were used to initiate infection of the Hu-mice, i) HTLV-1-WT which carries an intact HTLV-1 provirus genome, ii) HTLV-1-CTCF, which contains a provirus with a mutated vCTCF-BS which abolishes CTCF binding, and a stop codon immediate upstream of the mutated vCTCF-BS which deletes the last 23 amino acids of p12, and iii) HTLV-1-p12stop that contains the intact vCTCF-BS, but retains the same stop codon in p12 as in the HTLV-1-CTCF cell line. Hu-mice were infected with mitomycin treated or irradiated HTLV-1 producing cell lines. There was a delay in pathogenicity when Hu-mice were infected with the CTCF virus compared to mice infected with either p12 stop or WT virus. Proviral load (PVL), spleen weights, and CD4 T cell counts were significantly lower in HTLV-1-CTCF infected mice compared to HTLV-1-p12stop infected mice. Furthermore, we found a direct correlation between the PVL in peripheral blood and death of HTLV-1-CTCF infected mice. In cell lines, we found that the vCTCF-BS regulates Tax expression in a time-dependent manner. The scRNAseq analysis of splenocytes from infected mice suggests that the vCTCF-BS plays an important role in activation and expansion of T lymphocytes in vivo. Overall, these findings indicate that the vCTCF-BS regulates Tax expression, proviral load, and HTLV pathogenicity in vivo. Author SummaryHuman T-cell leukemia virus type 1 (HTLV-1) is a cause of leukemia and lymphoma, and several inflammatory medical disorders. The virus integrates in the host cell DNA, and it includes a single binding site for a cellular protein designated CTCF. This protein is important in regulation of many viruses, as well as properties of normal and malignant cells. In order to define the role of CTCF in HTLV-1 pathogenesis in vivo, we analyzed a mutant virus lacking the binding site in humanized mice. We found that this mutation slowed virus spread and attenuated the development of disease. Gene expression studies demonstrated a dynamic role of CTCF in regulating viral gene expression and T lymphocyte activation.

microbiology↗

bioRxiv2024

Personalized transcriptome signatures in a cardiomyopathy stem cell biobank

BACKGROUNDThere is growing evidence that pathogenic mutations do not fully explain hypertrophic (HCM) or dilated (DCM) cardiomyopathy phenotypes. We hypothesized that if a patients genetic background was influencing cardiomyopathy this should be detectable as signatures in gene expression. We built a cardiomyopathy biobank resource for interrogating personalized genotype phenotype relationships in human cell lines. METHODSWe recruited 308 diseased and control patients for our cardiomyopathy stem cell biobank. We successfully reprogrammed PBMCs (peripheral blood mononuclear cells) into induced pluripotent stem cells (iPSCs) for 300 donors. These iPSCs underwent whole genome sequencing and were differentiated into cardiomyocytes for RNA-seq. In addition to annotating pathogenic variants, mutation burden in a panel of cardiomyopathy genes was assessed for correlation with echocardiogram measurements. Line-specific co-expression networks were inferred to evaluate transcriptomic subtypes. Drug treatment targeted the sarcomere, either by activation with omecamtiv mecarbil or inhibition with mavacamten, to alter contractility. RESULTSWe generated an iPSC biobank from 300 donors, which included 101 individuals with HCM and 88 with DCM. Whole genome sequencing of 299 iPSC lines identified 78 unique pathogenic or likely pathogenic mutations in the diseased lines. Notably, only DCM lines lacking a known pathogenic or likely pathogenic mutation replicated a finding in the literature for greater nonsynonymous SNV mutation burden in 102 cardiomyopathy genes to correlate with lower left ventricular ejection fraction in DCM. We analyzed RNA-sequencing data from iPSC-derived cardiomyocytes for 102 donors. Inferred personalized co-expression networks revealed two transcriptional subtypes of HCM. The first subtype exhibited concerted activation of the co-expression network, with the degree of activation reflective of the disease severity of the donor. In contrast, the second HCM subtype and the entire DCM cohort exhibited partial activation of the respective disease network, with the strength of specific gene by gene relationships dependent on the iPSC-derived cardiomyocyte line. ADCY5 was the largest hubnode in both the HCM and DCM networks and partially corrected in response to drug treatment. CONCLUSIONSWe have a established a stem cell biobank for studying cardiomyopathy. Our analysis supports the hypothesis the genetic background influences pathologic gene expression programs and support a role for ADCY5 in cardiomyopathy.

genomics↗

bioRxiv2024

To be or not to be - Decoding the Trabecular Meshwork Cell Identity

The trabecular meshwork within the conventional outflow apparatus is critical in maintaining intraocular pressure homeostasis. In vitro studies employing primary cell cultures of the human trabecular meshwork (hTM) have conventionally served as surrogates for investigating the pathobiology of TM dysfunction. Despite its abundant use, translation of outcomes from in vitro studies to ex vivo and/or in vivo studies remains a challenge. Given the cell heterogeneity, performing single-cell RNA sequencing comparing primary hTM cell cultures to hTM tissue may provide important insights on cellular identity and translatability, as such an approach has not been reported before. In this study, we assembled a total of 14 primary hTM in vitro samples across passages 1-4, including 4 samples from individuals diagnosed with glaucoma. This dataset offers a comprehensive transcriptomic resource of primary hTM in vitro scRNA-seq data to study global changes in gene expression in comparison to cells in tissue in situ. We have performed extensive preprocessing and quality control, allowing the research community to access and utilize this public resource.

genomics↗

bioRxiv2024

A conserved lysine/arginine-rich motif in potyviral 6K1 protein is key in engaging autophagy-mediated self-degradation for completing pepper veinal mottle virus infection

Potyviruses possess one positive-sense single-stranded RNA genome mainly with polyprotein processing as their gene expression strategy. The resulting polyproteins are proteolytically processed by three virus-encoded proteases into 11 or 12 mature proteins. One of such, 6-kDa peptide 1 (6K1), is an understudied viral factor. Its function in viral infection remains largely mysterious. This study is to reveal part of its roles by using pepper veinal mottle virus (PVMV) as a model virus. Alanine substitution screening analysis revealed that 15 out of 17 conserved residues across potyviral 6K1 sequences are essential for PVMV infection. However, 6K1 protein is less accumulated in virus-infected cells, even though P3-6K1 junction is efficiently processed by NIa-Pro for its release, indicating that 6K1 undergoes a self-degradation event. Mutating the cleavage site to prevent NIa-Pro processing abolishes viral infection, suggesting that the generation of 6K1 along with its degradation might be important for viral multiplication. We corroborated that cellular autophagy is engaged in 6K1s degradation. Individual engineering of the 15 6K1 variants into PVMV was performed to allow for their expression along with viral infection. Five of such variants, D30A, V32A, K34A, L36A, and L39A, significantly interfere with viral infection. The five residues are enclosed in a conserved lysine/arginine-rich motif; four of them appear to be crucial in engaging autophagy-mediated self-degradation. Based on these data, we envisaged a scenario that potyviral 6K1s interact with an unknown anti-viral component to be co-degraded by autophagy to promote viral infection. IMPORTANCEPotyvirus is the largest genus of plant-infecting RNA viruses, which encompasses socio-economically important virus species, such as Potato virus Y, Plum pox virus, and Soybean mosaic virus. Like all picorna-like viruses, potyviruses express their factors mainly via polyprotein processing. Theoretically, viral factors P3 through CP, including 6K1, should share an equivalent number of molecules. The 6K1 is small in size ([~]6 kDa) and conserved across potyviruses, but less accumulated in virus-infected cells. This study demonstrates that cellular autophagy is engaged in the degradation of 6K1 to promote viral infection. In particular, we found a conserved lysine/arginine-rich motif in 6K1s across potyviruses that is engaged in this degradation event. This finding reveals one facet of a small protein that help understand the pro-viral role of cellular autophagy in viral infection.

microbiology↗

bioRxiv2024

A high-continuity and annotated reference genome of allotetraploid Siberian wildrye (Elymus sibiricus L., Poaceae: Triticeae)

Elymus sibiricus L. (Siberian wildrye, Es), a species belonging to the wheat tribe, is extensively employed as forage and for the reclamation of degraded grasslands within the Qinghai-Tibet Plateau (QTP). This study provides a high-quality reference genome assembly for the allotetraploid Es, which is composed of 14 pseudomolecules with the total genome size of 6.57 Gb. Our finding suggest that large-scale bursts of retrotransposons are critical for the genome expansion of Es. We discovered a translocation event between the Es4H and Es6H chromosomes with a low frequency of combination. Phylogenetic analyses of 90 Es accessions and 25 diploid accessions representing proposed ancestors from various habitats revealed the existence of four distinct populations. We further provided support for the hypothesis that the QTP is the center of origin and genetic diversity for Es. Collectively, our study offers valuable insights into the evolution of Es, as well as providing genomic resources for genetic enhancement in the Elymus genus and wheat tribe.

genomics↗

bioRxiv2024

Quinoline-based compounds can inhibit diverse enzymes that act on DNA

DNA methylation, as exemplified by cytosine-C5 methylation in mammals and adenine-N6 methylation in bacteria, is a crucial epigenetic mechanism driving numerous vital biological processes. Developing non-nucleoside inhibitors to cause DNA hypomethylation is a high priority, in order to treat a variety of significant medical conditions without the toxicities associated with existing cytidine-based hypomethylating agents. In this study, we have characterized fifteen quinoline-based analogs. Notably, compounds with additions like a methylamine (9) or methylpiperazine (11) demonstrate similar low micromolar inhibitory potency against both human DNMT1 (which generates C5-methylcytosine) and Clostridioides difficile CamA (which generates N6-methyladenine). Structurally, compounds 9 and 11 specifically intercalate into CamA-bound DNA via the minor groove, adjacent to the target adenine, leading to a substantial conformational shift that moves the catalytic domain away from the DNA. This study adds to the limited examples of DNA methyltransferases being inhibited by non-nucleotide compounds through DNA intercalation, following the discovery of dicyanopyridine-based inhibitors for DNMT1. Furthermore, our study shows that some of these quinoline-based analogs inhibit other enzymes that act on DNA, such as polymerases and base excision repair glycosylases. Finally, in cancer cells compound 11 elicits DNA damage response via p53 activation. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=108 SRC="FIGDIR/small/587980v1_ufig1.gif" ALT="Figure 1"> View larger version (34K): org.highwire.dtl.DTLVardef@da2e94org.highwire.dtl.DTLVardef@27f909org.highwire.dtl.DTLVardef@4ec628org.highwire.dtl.DTLVardef@1b384df_HPS_FORMAT_FIGEXP M_FIG C_FIG HighlightsO_LISix of fifteen quinoline-based derivatives demonstrated comparable low micromolar inhibitory effects on human cytosine methyltransferase DNMT1, and the bacterial adenine methyltransferases Clostridioides difficile CamA and Caulobacter crescentus CcrM. C_LIO_LICompounds 9 and 11 were found to intercalate into a DNA substrate bound by CamA. C_LIO_LIThese quinoline-based derivatives also showed inhibitory activity against various base excision repair DNA glycosylases, and DNA and RNA polymerases. C_LIO_LICompound 11 provokes DNA damage response via p53 activation in cancer cells. C_LI

molecular biology↗

bioRxiv2024

An interpretable integration model improving disease-free survival prediction for gastric cancer based on CT images and clinical parameters

Preoperative prediction of disease-free survival of gastric cancer is significantly important in clinical practice. Existing studies showed the potentials of CT images in identifying predicting the disease-free survival of gastric cancer. However, no studies to date have combined deep features with radiomics features and clinical features. In this study, we proposed a model which embedded radiomics features and clinical features into deep learning model for improving the prediction performance. Our models showed a 3%-5% C-index improvement and 10% AUC improvement in predicting DFS and disease event. Interpretation analysis including T-SNE visualization and Grad-CAM visualization revealed that the model extract biologically meaning features, which are potentially useful in predicting disease trajectory and reveal tumor heterogeneity. The embedding of radiomics features and clinical features into deep learning model could guide the deep learning to learn biologically meaningful information and further improve the performance on the DFS prediction of gastric cancer. The proposed model would be extendable to related problems, at least in few-shot medical image learning. Key PointsO_LIAn integration model combining deep features, radiomics features and clinical parameters improved disease-free-survival prediction of gastric cancer by 3%-5% C-index. C_LIO_LIEmbedding radiomics and clinical features into deep learning model through concatenation and loss design improved feature extraction ability of deep network. C_LIO_LIThe model revealed disease progression trajectory and tumor heterogeneity. C_LI

cancer biology↗

bioRxiv2024

How short peptides can disassemble ultra-stable tau fibrils extracted from Alzheimers disease brain by a strain-relief mechanism

Reducing fibrous aggregates of protein tau is a possible strategy for halting progression of Alzheimers disease (AD). Previously we found that in vitro the D-peptide D-TLKIVWC disassembles tau fibrils from AD brains (AD-tau) into benign segments with no energy source present beyond ambient thermal agitation. This disassembly by a short peptide was unexpected, given that AD-tau is sufficiently stable to withstand disassembly in boiling SDS detergent. To consider D peptide-mediated disassembly as a potential therapeutic for AD, it is essential to understand the mechanism and energy source of the disassembly action. We find assembly of D-peptides into amyloid-like fibrils is essential for tau fibril disassembly. Cryo-EM and atomic force microscopy reveal that these D-peptide fibrils have a right-handed twist and embrace tau fibrils which have a left-handed twist. In binding to the AD-tau fibril, the oppositely twisted D-peptide fibril produces a strain, which is relieved by disassembly of both fibrils. This strain-relief mechanism appears to operate in other examples of amyloid fibril disassembly and provides a new direction for the development of first-in-class therapeutics for amyloid diseases.

neuroscience↗