Biology subjects

zhang, y.

Publications and source records attributed to zhang, y..

At least 19 recordsLinked to original sources

bioRxiv2026

Localized Chromatin Rewiring Shapes Rhythmic Transcription within a Stable 3D Genome in Rice

How spatial genome architecture coordinates rhythmic gene transcription across scales in plants remains poorly understood. Here, we integrated bridge linker Hi-C (BL-Hi-C), H3K9ac ChIP-seq and RNA-seq at six time points across a 24-h circadian cycle in rice. While global higher-order structure was largely invariant, with 84.5% of A/B compartments and over 95% of topologically associating domains (TADs) conserved across all time points, we identified 557 dynamic compartment transitions coupled to changes in active chromatin marks. Within this stable structural scaffold, 35.8% of TADs containing clustered loci of multiple rhythmic genes with synchronized expression phases. In contrast, fine-scale intradomain contacts and chromatin loops exhibited pronounced temporal plasticity. Chromatin interactions connected rhythmic genes with diverse expression phases. Myb-binding motifs were enriched at chromatin interaction anchors at dawn and early day (ZT0 and ZT4). Together, our findings demonstrate that localized chromatin rewiring occurs within relatively stable architectural frameworks, providing a spatiotemporal context for circadian gene regulation in rice.

genomics↗

bioRxiv2026

Chromosome-level genome assembly of the Northeast China Brown Frog (Rana dybowskii)

Rana dybowskii is distributed across Northeast Asian and represents a valuable medical resource. A high-quality assembly of the genome has not yet been reproted. This species has 2n=24 chromosomes, but a huge genome size that estimated at 3.5 ~4.6 Gb in the previous studies. The relatively large chromosome size, exceeding hundreds of megabases, may result in difficulties of obtaining a complete chromosome level genome. Here, we constructed a chromosome-level genome assembly of R. dybowskii by integrating PacBio HiFi long-read sequencing for de novo assembly and CiFi (3C coupled with HiFi sequencing) for scaffolding. The final assembly consists of 12 chromosomes with a total of 3.95 Gb and a scaffold N50 length of 455 Mb. BUSCO assessment using the tetrapoda_odb12 database identified 94.2% complete and 0.5% fragmented orthologs, suggesting a high level of completeness of the assembly. Genomic annotation revealed that repetitive sequences comprise over 53% of the assembly, with retroelements and DNA transposons accounting for 22% and 25%, respectively. A total of 43,999 protein-coding genes were predicted with the assistance of RNA-seq reads from four tissues (muscle, eye, testis and skin). This high-quality chromosome-level reference genome provides a valuable genomic resource for advancing genetic studies of the species.

genomics↗

bioRxiv2026

The novel variant IBDV induces inflammatory responses via VP3-mediated pyroptosis by directly targeting Caspase-9

The novel variant infectious bursal disease virus (nVarIBDV), which has been widely prevalent since 2017, threatens the poultry industry by inducing severe bursal atrophy and intense inflammatory responses. Understanding the inflammatory mechanism underlying nVarIBDV infection is critical for preventing virus-induced damage. Pyroptosis, an inflammatory type of programmed cell death, may contribute to the inflammation and bursa of Fabricius damage induced by nVarIBDV. Here, we found that the nVarIBDV infection induced severe inflammatory responses both in vivo and in vitro, which were associated with pyroptosis. Further research revealed that the viral protein VP3 drove pyroptosis induced by nVarIBDV infection by indirectly activating the Caspase-3-GSDME pathway, leading to GSDME cleavage. Mechanistically, VP3 directly interacted with and activated Caspase-9, thereby initiating Caspase-9-Caspase-3-GSDME pathway. More importantly, the Serine 33 of VP3 was identified as the key amino acid for interacting with and activating Caspase-9. Mutation of this residue significantly weakened the ability of VP3 to interact with Caspase-9 and activate the Caspase-9-mediated pyroptosis pathway, altered the binding mode of the Caspase-9-VP3 protein complex, and ultimately reduced the ability of VP3 to induce pyroptosis. In conclusion, our results elucidated a novel mechanism by which nVarIBDV infection induced inflammatory responses whereby viral protein VP3 triggered pyroptosis by targeting the Caspase-9-Caspase-3-GSDME pathway. IMPORTANCEnVarIBDV is globally prevalent and poses a significant threat to the poultry industry by causing inflammation and damage to the bursa of Fabricius. This study revealed that nVarIBDV infection induced inflammatory responses and bursa of Fabricius damage, primarily mediated by the viral protein VP3 targeting Caspase-9 to induce pyroptosis. To our knowledge, this is the first report on the interaction between inflammatory responses induced by IBDV infection and pyroptosis. In addition, serine 33 was identified as the critical amino acid required for maintaining the stability of the interaction between VP3 and Caspase-9 and activation of the Caspase-9-Caspase-3-GSDME cascade pyroptosis pathway and further inflammatory responses. Collectively, our findings not only elucidate a novel mechanism by which nVarIBDV induces pyroptosis and further leads to inflammation but also suggest that the identification of this serine site may represent a potential therapeutic target to block nVarIBDV-induced inflammation and bursa of Fabricius damage.

microbiology↗

bioRxiv2025

The Notch/snRNA negative feedback regulation and its implications in Alzheimer's disease

Alzheimers disease (AD) is a multifactorial disorder whose hallmark lesions have been recognized for more than a century, yet its molecular pathobiology remains fragmentary. During our AD research, we discovered a direct mechanistic axis that links the Notch signalling pathway -- a crucial regulator of cell differentiation and proliferation--with the small nuclear RNAs (snRNAs) that drive pre-mRNA splicing. For the first time, our discovery unifies two previously siloed paradigms -- cell signal transduction and RNA splicing, expanding the fundamental knowledge of gene expression and its regulation. The first contribution of the present study is the discovery and validation of the Notch/snRNA negative feedback regulation. More implications include: (1) it reframes splicing-related diseases as dynamic consequences of pathway dysregulation rather than static snRNA-gene mutations; (2) it explains why global splicing abnormalities appear in AD, aging, and cancer even when snRNA genes themselves are not mutated; and (3) it provides an immediate, druggable "splicing rheostat" for any Notch-hyperactive disorder, such as AD.

cell biology↗

bioRxiv2025

Protenix - Advancing Structure Prediction Through a Comprehensive AlphaFold3 Reproduction

In this technical report, we present Protenix, a comprehensive reproduction of AlphaFold3 (AF3), aimed at advancing the field of biomolecular structure prediction. Protenix tackles the challenges of predicting complex interactions involving proteins, ligands, and nucleic acids, while enhancing accessibility and reproducibility. Across diverse benchmarks, including PoseBusters V2, low-homology PDB sets, and CASP15 RNA, Protenix achieves state-of-the-art performance in protein-ligand, protein-protein, and protein-nucleic acid predictions. We also address limitations, such as potential memorization effects, and outline future directions for improvement. By open-sourcing Protenix, we aim to democratize advanced structure prediction tools and accelerate interdisciplinary research in computational biology and drug discovery.

bioinformatics↗

bioRxiv2024

A phosphoramidate modification of FUDR, NUC-3373, causes DNA damage and DAMPs release from colorectal cancer cells, potentiating lymphocyte-induced cell death

Colorectal cancer (CRC) is one of the leading causes of cancer-related mortality worldwide with 5-FU still the primary chemotherapeutic of choice. With the increasing use of immunotherapies, much research is focused on the ability to make tumours more immunogenic, thereby rationalising combination with immunotherapy. Here we investigate whether NUC-3373, a phosphoramidate transformation of FUDR, enhances immunogenicity in CRC cell lines and facilitates lymphocyte mediated cell death in vitro. At sub IC50 doses NUC-3373 upregulates damage associate molecular patterns (DAMPs) in both HCT116 and SW480 cells and increases surface expression of MHCII and PD-L1. Pre-treatment with NUC-3373 and subsequent coculture with NK-92 MI natural killer cells caused an increase in LAMP1 expression (degranulation), production of IFN-{psi}, and NK-mediated cytotoxicity compared to vehicle controls. Cocultures with patient-derived PBMCs with heterologous CRC cells pre-treated with NUC-3373 demonstrated increased cell death compared to both vehicle controls and monocultures of CRC cells exposed to NUC-3373. Lastly, the PD-1 immune checkpoint inhibitor nivolumab showed synergistic activity when HCT116 cells were pre-treated with NUC-3373. To conclude, we show that NUC-3373 can modulate immune signaling and may therefore facilitate immune mediated tumour cell death.

cancer biology↗

bioRxiv2024

Comparative metabolomics of wild and cultivated yew: insights into metabolic diversity and environmental adaptability

Taxus cuspidata is a well-known gymnosperm with great ornamental and medicinal value. The study aims to reveal the metabolic differences between wild and cultivated species of T. cuspidata, and to analyze the genetic and ecological factors behind these differences. This study conducted a comparative metabolomics analysis of wild and cultivated T. cuspidata based on LC-MS/MS technology. The results showed: (1) A total of 7030 metabolites were identified, primarily including flavonoids, organic acids, phenolic acids, amino acids and their derivatives, lipids, and alkaloids, among others; (2) 2381 differential metabolites were confirmed: 949 higher in cultivated species and 1432 higher in wild species, which suggesting wild species have more advantageous metabolites. (3) Through KEGG annotation, 20 significant metabolic pathways were identified, with sugar metabolism, photosynthesis, and sulfur metabolism were significantly higher in wild species. In contrast, pathways related to most amino acids and their derivatives, flavonoids, terpenoids, polyphenols, alkaloids, and plant hormones were significantly higher in cultivated species. In conclusion, the wild species are inferior to cultivated species in terms of drought resistance, growth rate, and medicinal value for cancer treatment. However, the advantage of wild species lies in their significantly higher diversity of upregulated secondary metabolites due to their rich genetic resources, which exhibit greater metabolic diversity and environmental adaptability, which enhances their survival capability. Furthermore, the wild species exhibit significantly higher efficiency in photosynthesis and sugar metabolism compared to cultivated species, enabling them to better withstand low-temperature stress. This study has provided the direction for research into the mechanisms of stress resistance in T. cuspidata and has also established a theoretical basis for formulating scientifically and reasonable conservation strategies for wild species.

molecular biology↗

bioRxiv2024

Prediction of binding sites of GPCRs based on 3D convolutional neural networks

G protein-coupled receptors are a class of receptor proteins located on the cell membrane, regulating the perception and response of cells to various external signals. Identifying the binding sites of G protein-coupled receptors plays a crucial role in understanding their allosteric modulation mechanisms. However, obtaining the crystal structure of the complex through experimental means and subsequently identifying the binding sites require substantial resources. With the development of computer-aided computation, deep learning can effectively predict the binding sites between proteins and ligands. This study predicted the binding sites of G protein-coupled receptors based on 3D convolutional neural network. A total of 108 G protein-coupled receptors recorded in the scPDB database were collected for this study, and a 3D convolutional neural network model was established based on these three-dimensional structures. Firstly, the PDB file of the protein is voxelized and segmented into different channels according to the type of atoms in a certain region. Then, 3D convolutional neural network is employed to predict the binding site through traversal, and the optimal voxel box size and model parameters are determined based on performance evaluation metrics. The established 3D convolutional neural network accurately predicts the binding site of G protein-coupled receptors, with an accuracy as high as 0.942, precision of 0.678, and recall rate of 0.532. Additionally, using the backpropagation algorithm, the gradients of the input data are calculated, and the importance of input elements on the final classification result is analyzed.

bioinformatics↗

bioRxiv2024

Adipocyte microRNA-802 promotes adipose tissue inflammation and insulin resistance by modulating macrophages in obesity

Adipose tissue inflammation is now considered to be a key process underlying metabolic diseases in obese individuals. However, it remains unclear how adipose inflammation is initiated and maintained or the mechanism by which inflammation develops. We found that microRNA-802 (miR-802) expression in adipose tissue is progressively increased with the development of dietary obesity in obese mice and humans. The increasing trend of miR-802 preceded the accumulation of macrophages. Adipose tissue-specific knockout of miR-802 lowered macrophage infiltration and ameliorated systemic insulin resistance. Conversely, the specific overexpression of miR-802 in adipose tissue aggravated adipose inflammation in mice fed a high-fat diet. Mechanistically, miR-802 activates noncanonical and canonical NF-{kappa}B pathways by targeting its negative regulator, TRAF3. Next, NF-{kappa}B orchestrated the expression of chemokine and SREBP1, which translated into strong recruitment and M1-like polarization of macrophages. Our findings indicate that miR-802 endows adipose tissue with the ability to recruit and polarize macrophages, which underscores miR-802 as an innovative and attractive candidate for miRNA-based immune therapy for adipose inflammation.

immunology↗

bioRxiv2024

Student-teacher inter-brain coupling causally predict academic achievement over semesters

Student performance exhibits dynamic fluctuates throughout the learning process, which reflects deeper educational myths and impacts educational quality and equity. Although teachers are believed to play a pivotal role in shaping these changes, the mechanisms through which student-teacher interactions drive performance variations across individual students remain unclear. The present study investigates student-teacher inter-brain coupling in authentic classroom environments through longitudinal neurophysiological recordings spanning three academic semesters. Using wearable hyperscanning technology, a total of 4,175 longitudinal EEG recordings were collected from 107 junior and senior students across 393 regular Chinese and math classes. While static models revealed a negative association between student-teacher inter-brain coupling and math performance--likely reflecting teacher selection effects--dynamic analyses showed that increased coupling, particularly in the high-beta frequency band, causally predicted academic improvement in both subjects. These findings highlight the functional significance of inter-brain coupling in real-world classroom learning, providing robust ecological validity and supporting its potential as a measurable marker of effective pedagogy. Significance StatementThis study is the first to provide causal evidence, based on real-world classroom data, that enhanced student-teacher inter-brain coupling improved academic performance. Leveraging 4,175 longitudinal EEG recordings from 107 students across three semesters, we show that higher coupling levels predict better learning outcomes over time. Capturing neural dynamics in authentic classrooms, this research offers high ecological validity and strong potential for practical application and large-scale educational implementation.

neuroscience↗

bioRxiv2024

Inferring gene regulatory networks by hypergraph variational autoencoder

In constructing Gene Regulatory Networks (GRNs), it is crucial to consider cellular heterogeneity and differential gene regulatory modules. However, traditional methods have predominantly focused on cellular heterogeneity, approaching the subject from a relatively narrow scope. We present HyperG-VAE, a Bayesian deep generative model that utilizes a hypergraph to model single-cell RNA sequencing (scRNA-seq) data. HyperG-VAE employs a cell encoder with a Structural Equation Model to address cellular heterogeneity and build GRNs, alongside a gene encoder using hypergraph self-attention to identify gene modules. Encoders are synergistically optimized by a decoder, enabling HyperG-VAE to excel in GRN inference, single-cell clustering, and data visualization, evidenced by benchmarks. Additionally, HyperG-VAE effectively reveals gene regulation patterns and shows robustness in varied downstream analyses, demonstrated using B cell development data in bone marrow. The interplay of encoders by the overlapping genes between predicted GRNs and gene modules is further validated by gene set enrichment analysis, underscoring that the gene encoder boosts the GRN inference. HyperG-VAE proves efficient in scRNA-seq data analysis and GRN inference.

bioinformatics↗

bioRxiv2024

Collective movement of schooling fish reduces locomotor cost in turbulence

The ecological and evolutionary benefits of collective behaviours are rooted in the physical principles and physiological mechanisms underpinning animal locomotion. We propose a turbulence sheltering hypothesis that collective movements of fish schools in turbulent flow can reduce the total energetic cost of locomotion by shielding individuals from the perturbation of chaotic turbulent eddies. We test this hypothesis by quantifying energetics and kinematics in schools of giant danio (Devario aequipinnatus) compared to solitary individuals swimming under control and turbulent conditions over a wide speed range. We discovered that, when swimming at high speeds and high turbulence levels, fish schools reduced their total energy expenditure (TEE, both aerobic and anaerobic energy) by 63-79% compared to solitary fish. Solitary individuals spend [~]25% more kinematic effort (tail beat amplitude*frequency) to swim in turbulence at higher speeds than in control conditions. However, fish schools swimming in turbulence reduced their three-dimensional group volume by 41-68% (at higher speeds) and did not alter their kinematic effort compared to control conditions. This substantial energy saving highlighted a [~]261% higher TEE when fish swimming alone in turbulence are compared to swimming in a school. Schooling behaviour could mitigate turbulent disturbances by sheltering fish within schools from the eddies of sufficient kinetic energy that can disrupt the locomotor gaits. Providing a more desirable internal hydrodynamic environment could be one of the ecological drivers underlying collective behaviours in a dense fluid environment. One-Sentence SummaryThe collective movement of fish schools substantially reduces the energetic cost of locomotion in turbulence compared to that of swimming alone.

physiology↗

bioRxiv2024

Elucidation of neuronal activity in mouse models of TMJ injury by in vivo GCaMP Ca2+ imaging of intact trigeminal ganglion neurons

Patients with temporomandibular disorders (TMD) typically experience facial pain and discomfort or tenderness in the temporomandibular joint (TMJ), causing disability in daily life. Unfortunately, existing treatments for TMD are not always effective, creating a need for more advanced, mechanism-based therapies. In this study, we used in vivo GCaMP3 Ca2+ imaging of intact trigeminal ganglia (TG) to characterize functional activity of the TG neurons in vivo, specifically in TMJ animal models. This system allows us to observe neuronal activity in intact anatomical, physiological, and clinical conditions and to assess neuronal function and response to various stimuli. We observed a significant increase in spontaneously and transiently activated neurons responding to mechanical, thermal, and chemical stimuli in the TG of forced mouth open (FMO) mice. An inhibitor of the CGRP receptor significantly attenuated FMO-induced facial hypersensitivity. In addition, we confirmed the attenuating effect of CGRP antagonist on FMO-induced sensitization by in vivo GCaMP3 Ca2+ imaging of intact TG. Our results contribute to unraveling the role and activity of TG neurons in the TMJ pain animal models of TMD, bringing us closer understanding the pathophysiological processes underlying TMD. Our study also illustrates the utility of in vivo GCaMP3 Ca2+ imaging of intact TG for studies aimed at developing more targeted and effective treatments for TMD.

neuroscience↗

bioRxiv2023

Omics-assisted systematic exploration of the intricate regulatory network of guvermectin biosynthesis centered by the cluster-situated regulator GvmR in Streptomyces

Guvermectin, produced by Streptomyces bacteria, is a purine nucleoside natural product recently registered as a new biopesticide to boost rice yield. Despite its importance, the regulatory network governing guvermectin biosynthesis remains largely unknown, severely impeding industrial-scale production and widespread application in rice production. Here, we investigated the diverse regulatory mechanisms employed by the cluster-situated regulatory gene, gvmR, in controlling guvermectin production from the perspective of widespread disturbance of gene expression at genome scale. GvmR activates the expression of guvermectin cluster by binding to the gvmR, gvmA and O1 promoters. Additionally, GvmR binds to the promoter of scnR1, a previously unidentified and highly conserved regulator in Streptomyces. scnR1 overexpression significantly suppressed guvermectin production by regulating the guvermectin cluster through binding to the same promoters as GvmR. Transcriptomic analysis revealed that GvmR extensively influences the expression of numerous genes located outside the guvermectin cluster, including the precursor supply (purine biosynthesis) and energy synthesis (oxidative phosphorylation) pathway genes, as well as 252 transcriptional regulatory genes. By genetic screening from 48 of these 252 regulatory genes, we identified additional five highly conserved genes that impact guvermectin production, suggesting a functional interplay between GvmR and highly conserved regulators in coordinating guvermectin production. These findings enrich our knowledge of the regulatory network governing guvermectin biosynthesis and offer a broadly applicable approach for investigating the molecular regulation of natural product biosynthesis and their high-titer production.

microbiology↗

bioRxiv2023

Sparse Modeling of Genomic Landscape Identifies Pathogenic Processes and Therapeutic Targets in Metastatic Breast Cancer

Breast cancer is a heterogeneous disease and ranks as one of the most lethal and frequently detected disease in the world. It poses significant challenges for precision therapy. To better decipher the patterns of heterogeneous nature in human genome and converge them into common functionals, mutational signatures are introduced to define the types of DNA damage, repair and replicative mechanisms that shape the genomic landscape of each cancer patient. In this study, we developed a deep learning (DL) model, MetaWise 2.0, based on pruning technology that improved model generalization with deep sparsity. We applied it to patient samples from multiple sequencing studies, and identified statistically significant mutational signatures associated with metastatic progression using Shapley additive explanations (SHAP). We also employed gene cumulative contribution abundance analysis to link the mutational signatures with relevant genes, which could unearth the shared molecular mechanisms behind tumorigenesis and metastasis of each patient and lead to novel therapeutic target identification. Our study illustrates that MetaWise 2.0 is an effective DL tool for discovering clinically meaningful mutational signatures in metastatic breast cancer (MBC) and relating them directly to relevant biological functions and gene targets. These findings could facilitate the development of novel therapeutic strategies and improve the clinical outcomes for individual patients.

bioinformatics↗

bioRxiv2023

Patient-Specific Analysis of Co-expression Networks for Predicting Clinical Outcomes in Breast Cancer

To effectively understand the underlying mechanisms of disease and inform the development of personalized therapies, it is critical to harness the power of differential co-expression network analysis. Despite the promise of differential co-expression network analysis in precision medicine, current approaches have a major limitation: they measure an average differential network across multiple samples, which means the specific etiology of individual patients is often overlooked. To address this, we present Cosinet, a differential co-expression-based single-sample network rewiring degree quantification tool. By analyzing two breast cancer datasets, we demonstrate that Cosinet can identify important differences in gene co-expression patterns between individual patients and generate scores for each individual that are significantly associated with overall survival, recurrence-free interval, and other clinical outcomes, even after adjusting for risk factors such as age, tumor size, HER2 status and PAM50 subtypes. Cosinet represents a remarkable development toward unlocking the potential of differential co-expression analysis in the context of precision medicine.

bioinformatics↗

bioRxiv2023

The Multifaceted Effects of Short-Term Acute Hypoxia Stress: Insights into the tolerance mechanism of Propsilocerus akamusi (Diptera: Chironomidae)

Plenty of freshwater species, especially macroinvertebrates that are essential to the provision of numerous ecosystem functions, encountered higher mortality due to acute hypoxia. However, within the family Chironomidae, a wide range of tolerance to hypoxia/anoxia is displayed and Propsilocerus akamusi depends on this great endurance to become a dominant species in eutrophic lakes. To further understand how P. akamusi responds to acute hypoxic stress, we used multi-omics analysis in combination with histomorphological characteristics and physiological indicators. To evaluate enzyme activity, the transcriptome and metabolome, and histomorphological characteristics, we set up two groups: a control group (DO 8.4mg/L) and a hypoxic group (DO 0.39mg/L). With blue-black chromatin, cell tightness, cell membrane invagination, and the production of apoptotic vesicles, tissue cells displayed typical apoptotic features. While lactate dehydrogenase (LDH), (Alcohol dehydrogenase) ADH, catalase (CAT), and Na+/K+ -ATPase (NKA) activities were dramatically enhanced under hypoxic stress, glycogen content, and superoxide dismutase (SOD) activities were significantly reduced compared to the control group. The above results were further supported by the joint analysis of the transcriptome and metabolome, which further revealed that in addition to carbohydrates, including glycogen, the energy metabolism of the fatty acid, trehalose, and glyoxylate cycles is also included. Furthermore, we also revealed the ethanol tested in hypoxic stress should derive from symbiodinium of P akamusi. Understanding the processes which enable P. akamusi to survive lengthy periods of hypoxia in eutrophic lakes might help to design sensitive biomonitoring procedures, and this species has the potential to be used as an effective eutrophication indicator.

biophysics↗

bioRxiv2022

Structure-Kinetic Relationship for Drug Design Revealed by PLS Model with Retrosynthesis-Based Pre-trained Molecular Representation and Molecular Dynamics Simulation

Drug design based on their molecular kinetic properties is growing in application. Pre-trained molecular representation based on retrosynthesis prediction model (PMRRP) was trained from 501 inhibitors of 55 proteins and successfully predicted the koff values of 38 inhibitors for HSP90 protein from an independent dataset. Our PMRRP molecular representation outperforms others such as GEM, MPG, and common molecular descriptors from RDKit. Furthermore, we optimized the accelerated molecular dynamics to calculate relative retention times for 128 inhibitors of HSP90. We observed high correlation between the simulated, predicted, and experimental -log(koff) scores. Combining machine learning (ML) and molecular dynamics (MD) simulation help design a drug with specific selectivity to the target of interest. Protein-ligand interaction fingerprints (IFPs) derived from accelerated MD further expedite the design of new drugs with the desired kinetic properties. To further validate our koff ML model, from the set of potential HSP90 inhibitors obtained by similarity search of commercial databases, we identified two novel molecules with better predicted koff values and longer simulated retention time than the reference molecules. The IFPs of the novel molecules with the newly discovered interacting residues along the dissociation pathways of HSP90 shed light on the nature of the selectivity of HSP90 protein. We believe the ML model described here is transferable to predict koff of other proteins and enhance the kinetics-based drug design endeavor.

biophysics↗