Biology subjects

Guo, J.

Publications and source records attributed to Guo, J..

At least 181 records · Page 10Linked to original sources

bioRxiv2024

Mitigation of Fetal Irradiation Injury from Mid-Gestation Total Body Radiation with Mitochondrial-Targeted GS-Nitroxide JP4-039

Victims of a radiation terrorist event will include pregnant women and unborn fetuses. Mitochondrial dysfunction and oxidative stress are key pathogenic factors of fetal irradiation injury. The goal of this preclinical study is to investigate the efficacy of mitigating fetal irradiation injury by maternal administration of the mitochondrial-targeted gramicidin S (GS)- nitroxide radiation mitigator, JP4-039. Pregnant female C57BL/6NTac mice received 3 Gy total body ionizing irradiation (TBI) at mid-gestation embryonic day 13.5 (E13.5). Using novel time- and-motion-resolved 4D in utero magnetic resonance imaging (4D-uMRI), we found TBI caused extensive injury to the fetal brain that included cerebral hemorrhage, loss of cerebral tissue, and hydrocephalus with excessive accumulation of cerebrospinal fluid (CSF). Histopathology of the fetal mouse brain showed broken cerebral vessels and elevated apoptosis. Further use of novel 4D Oxy-wavelet MRI capable of probing in vivo mitochondrial function in intact brain revealed significant reduction of mitochondrial function in the fetal brain after 3Gy TBI. This was validated by ex vivo Oroboros mitochondrial respirometry. Maternal administration JP4-039 one day after TBI (E14.5), which can pass through the placental barrier, significantly reduced fetal brain radiation injury and improved fetal brain mitochondrial respiration. This also preserved cerebral brain tissue integrity and reduced cerebral hemorrhage and cell death. As JP4-039 administration did not change litter sizes or fetus viability, together these findings indicate JP4-039 can be deployed as a safe and effective mitigator of fetal radiation injury from mid-gestational in utero ionizing radiation exposure. One Sentence SummaryMitochondrial-targeted gramicidin S (GS)-nitroxide JP4-039 is safe and effective radiation mitigator for mid-gestational fetal irradiation injury.

developmental biology↗

bioRxiv2024

Mapping protein-metabolite interactions in E. coli by integrating chromatographic techniques and co-fractionation mass spectrometry.

In our pursuit of understanding the protein-metabolite interactome, we introduced PROMIS, a co-fractionation mass spectrometry (CF-MS) technique focusing on biosynthetic and regulatory processes. However, the challenge lies in distinguishing true interactors from coincidental co-elution when a metabolite co-fractionates with numerous proteins. To address this, we integrated two chromatographic techniques-- size exclusion and ion exchange--to enhance the mapping of protein-metabolite interactions (PMIs) in Escherichia coli. This integration aims to refine the PMI network by considering size and charge characteristics, resulting in 994 interactions involving 51 metabolites and 465 proteins. The PMI network is enriched for known and predicted interactions validating our approachs efficacy. Furthermore, the analysis of protein targets for different metabolites revealed novel functional insights, such as the connection between proteinogenic dipeptides and fatty acid biosynthesis. Notably, we uncovered an inhibitory interaction between the riboflavin degradation product lumichrome and orotate phosphoribosyltransferase (PyrE), a key enzyme in de novo pyrimidine synthesis. Lumichrome supplementation mimicked the biofilm formation inhibition observed in a{Delta} pyrE mutant strain, suggesting lumichrome role in integrating pyrimidine and riboflavin metabolism with quorum sensing and biofilm formation. In summary, our integrated chromatographic approach significantly advances PMI mapping, offering novel insights into functional associations and potential regulatory mechanisms in E. coli.

biochemistry↗

bioRxiv2024

In-silico Drug Repurposing pipeline for Epilepsy: Integrating Deep Learning and Structure-based Approaches

Due to considerable global prevalence and high recurrence rate, the pursuit of effective new medication for epilepsy treatment remains an urgent and significant challenge. Drug repurposing emerges as a cost-effective and efficient strategy to combat this disorder. This study leverages the transformer-based deep learning methods coupled with molecular binding affinity calculation to develop a novel in-silico drug repurposing pipeline for epilepsy. The number of candidate inhibitors against 24 target proteins encoded by gain-of-function (GOF) genes implicated in epileptogenesis ranged from zero to several hundreds. Our pipeline has repurposed the medications with most anti-epileptic drugs (AEDs) and nearly half psychiatric medications, highlighting the effectiveness of our pipeline. Furthermore, Lomitapide, a cholesterol-lowering drug, first emerged as particularly noteworthy, exhibiting high binding affinity for 10 targets and verified by molecular dynamics (MD) simulation and mechanism analysis. These findings provided a novel perspective on therapeutic strategies for other central nervous system (CNS) disease.

pharmacology and toxicology↗

bioRxiv2024

Turnover of RNA-binding Proteins and MicroRNAs by intrinsically disordered region-directed ZSWIM8 ubiquitin ligase during brain development

Widely present in mammalian proteomes, intrinsically disordered regions (IDRs) in proteins play important biological functions by conferring structural flexibility and mediating biomolecular interactions. IDR-containing proteins, including many RNA-binding proteins (RBPs), are prone to misfolding and aggregation and must be constantly monitored. Here we show that the conserved ZSWIM8-type Cullin-RING ubiquitin ligase (CRLZSWIM8) is a master regulator of such proteins during brain development. ZSWIM8 selects its substrates via an IDR-dependent mechanism, and deletion of ZSWIM8 causes aberrant accumulation of numerous RBPs including AGO2 and ELAV1 in neonatal brains. Furthermore, AGO2 ubiquitination by ZSWIM8 is triggered by microRNA binding, leading to target-directed microRNA degradation (TDMD) of MiR7. Dysregulation of MiR7 in the absence of ZSWIM8 results in defects in oligodendrocyte maturation and functions. Together, our findings have demonstrated that, by utilizing variable target-recognition strategies, ZSWIM8 controls the abundance of conformationally flexible RBPs and miRNA metabolism that are essential for brain development. TeaserA conserved ubiquitin ligase controls the quality of disordered proteins to ensure brain development.

developmental biology↗

bioRxiv2024

Cell Type Specific Responses of the Endoplasmic Reticulum Dynamics to Environmental Stress

To overcome the challenge of balancing imaging speecd and resolution, which currently limits the accurate identification of structural and dynamic changes in the study of endoplasmic reticulum (ER) in plant cells. This research employs structured illumination microscopy techniques to achieve super-resolution real-time imaging of the ER in live Arabidopsis materials. Additionally, a self-supervised denoising framework (Blind2Unblind) was optimized to further enhance the signal-to-noise ratio of rapid microscopic imaging. Based on the images with high quality, a method for quantitative analysis of ER structures using time-lapse images is developed. Moreover, detections of changes in ER structures under environmental stress are conducted to verify the effectiveness of the method. Moreover, correlation analyses of various parameters indicate a significant positive correlation between the area and length of tubular ER with the number of growth tips and tricellular junctions, while the area of ER cisternae and bulk flow exhibits a significant negative correlation with the area and length of tubules. The super-resolution imaging and dynamic analysis method developed in this study will provide new technical approaches for further elucidating the function and regulatory mechanisms of the plant ER.

plant biology↗

bioRxiv2024

Engineered minimal type I CRISPR-Cas system for transcriptional activation and base editing in human cells

Type I CRISPR-Cas systems are widespread and have exhibited remarkable versatility and efficiency in genome editing and gene regulation in prokaryotes. However, due to the multi-subunit composition and large size, their application in eukaryotes has not been thoroughly investigated. Here, we demonstrate that the type I-F2 Cascade, the most compact among type I systems and significantly smaller than SpCas9, can be developed into programmable tools for use in human cells. For transcriptional activation, the efficiency of the tool based on the engineered I-F2 system can match or surpass that of dCas9. Besides, narrow editing windows limit the application of base editors. Although the R-loop formed by Cascade is much wider than that by Cas9 or Cas12, the potential of base editing with Cascade has not yet been explored. We successfully created a base editor with the I-F2 Cascade, which induces a considerably wide editing window ([~]30 nt) with a bimodal distribution. The wide editing window can expand the range of targetable sites and can be useful for disrupting functional sequences and genetic screening. The editing efficiency can achieve 50% in human cells. This research underscores the application potential of compact type I systems in eukaryotes and developed a new base editor with an extraordinary wide editing window.

molecular biology↗

bioRxiv2024

LRG1 promotes atherosclerosis by activating macrophages

BackgroundAtherosclerosis is a chronic inflammatory disease of the arterial wall characterized by the accumulation of cholesterol-rich lipoproteins in macrophages. Leucine-rich alpha-2 glycoprotein 1 (LRG1) is a circulating protein associated with inflammation, however, its role in atherosclerosis remains unclear. This study identified its role in macrophage pro-inflammatory differentiation and revealed the relationship between LRG1 and atherosclerosis. MethodWe evaluated the impact of LRG1 on atherosclerosis progression by analyzing atherosclerotic tissue and serum samples from patients with coronary artery disease (CAD) and healthy individuals and analyzed its role in such a process using two types of mice models: Apoe knock-out mice (Apoe-/-) and Apoe and Lrg1 double knock-out mice (Apoe-/-/Lrg1-/-). These mice were fed with a high-fat diet for 16 to 32 weeks to simulate conditions exacerbating atherosclerosis. To examine the effects of inhibiting LRG1 on atherogenesis, we administered intraperitoneal injections of LRG1 neutralizing antibody (50g/kg) weekly to Apoe-/- mice for 8 weeks. We conducted in vitro assays using bone marrow-derived macrophages isolated from wild-type mice and analyzed transcriptional signatures using RNA sequencing. Additionally, we utilized small molecular inhibitors to validate the signaling pathway through which LRG1 promotes macrophage-driven inflammation. ResultsLRG1 levels were found to be elevated in patients with atherosclerosis and correlated with higher levels of a plasma pro-inflammatory biomarker high-sensitive C-reactive protein (hsCRP), and several macrophage-related pro-inflammatory markers including CD68, VE-Cadherin and VCAM-1. In a high fat diet induced Apoe-/- mouse atherosclerosis model, the deletion of LRG1 gene significantly delayed atherogenesis progression and reduced levels of macrophage-related pro-inflammatory cytokines. Addition of purified LRG1 to cultured macrophages stimulated those macrophages to pro-inflammatory M1-like polarization regulated by the activation of ERK and JNK pathways. An anti-LRG1 neutralizing antibody effectively blocked LRG1-induced macrophage M1-like polarization in vitro and conferred therapeutic benefits to animals with ApoE deficiency-induced atherosclerosis. ConclusionLRG1 plays an important pro-inflammatory role in atherosclerosis by influencing macrophage polarization towards a pro-inflammatory state. The inhibition of LRG1 with neutralizing antibodies may offer a potential therapeutic strategy for patients with atherosclerosis by mitigating the pro-inflammatory response and delaying disease progression, offering a novel therapy in atherosclerosis management. Translational PerspectiveAtherosclerosis, a persistent inflammatory condition affecting the arterial wall, serves as the underlying pathophysiological basis for acute ischemic cardiovascular events. The involvement of macrophages is crucial in the advancement of atherosclerosis. In this investigation, heightened levels of plasma LRG1 were observed in individuals with coronary artery disease. Moreover, this study presents initial evidence highlighting LRG1 as a pivotal activator of macrophages, instigating a pro-inflammatory M1 polarization during atherogenesis through the activation of ERK1/2 and JNK pathways. The use of an anti-LRG1 neutralizing antibody demonstrated a delay in atherosclerosis progression in an animal model, suggesting a potential therapeutic target for atherosclerosis treatment. Suppression of LRG1 production could impede atherosclerosis advancement and enhance plaque stability. Utilizing neutralizing antibodies against LRG1 emerges as a promising therapeutic approach for treating atherosclerosis.

pathology↗

bioRxiv2024

Early evolution of the ecdysozoan body plan

Extant ecdysozoans (moulting animals) are represented by a great variety of vermiform or articulated organisms. However, controversies remain about the nature of their ancestral body plan although the vermiform hypothesis seems to prevail. We describe here Beretella spinosa gen et sp. nov. a tiny ecdysozoan from the early Cambrian, Yanjiahe Formation, South China, with an unusual sack-like appearance, single opening, and spiny ornament. Beretella has no equivalent among animals, except Saccorhytus from the basal Cambrian. Phylogenetic analyses resolve both forms as a sister group (Saccorhytida) to all known Ecdysozoa, thus suggesting that ancestral ecdysozoans may have been non-vermiform animals. Saccorhytids are likely to represent an early dead-end off-shot along the stem-line Ecdysozoa that possibly evolved through anatomical simplification (e.g. lack of anus). Although extinct during the Cambrian, this animal lineage provides precious insight into the early evolution of Ecdysozoa and the nature (possibly non-vermiform) of the earliest representatives of the group.

paleontology↗

bioRxiv2024

Root-specific secondary metabolism at the single-cell level: a case study of theanine metabolism and regulation in the roots of tea plants (Camellia sinensis)

Root-synthesized secondary metabolites are critical quality-conferring compounds of foods, plant-derived medicines, and beverages. However, information at a single-cell level on root-specific secondary metabolism remains largely unexplored. L-theanine, an important quality component of tea, is primarily synthesized in roots, from which it is then transported to new shoots of tea plant. In this study, we present a single-cell RNA sequencing (scRNA-seq)-derived map for the tea plant root, which enabled cell-type-specific analysis of glutamate and ethylamine (two precursors of theanine biosynthesis) metabolism, and theanine biosynthesis, storage, and transport. Our findings support a model in which the theanine biosynthesis pathway occurs via multicellular compartmentation and does not require high co-expression levels of transcription factors and their target genes within the same cell cluster. This study provides novel insights into theanine metabolism and regulation, at the single-cell level, and offers an example for studying root-specific secondary metabolism in other plant systems.

plant biology↗

bioRxiv2024

Liver cancer initiation is dependent on metabolic zonation but decoupled from premalignant clonal expansion

The origin of cancer is poorly understood because cells that obtain truncal mutations are rarely fate mapped in their native environments. A defining feature of the liver is zonation, or the compartmentalization of metabolic functions in hepatocytes located in distinct regions of the lobule 1. However, it is unknown if cancers develop in some zones but not others, and if there are metabolic determinants of cancer risk that track with cellular position. To study cancer initiation, we examined the effect of activating mutations in Ctnnb1 and loss of function mutations in Arid2, two of the most commonly co-mutated genes in hepatocellular carcinoma (HCC) 2. We exploited glutamine synthetase (GS) as a faithful fate mapping marker of Ctnnb1 mutant hepatocytes. By introducing mutations in distinct zones in a mosaic fashion, we showed that position and metabolic context regulate clone expansion. Mutant clones were maintained in zone 1 but largely outcompeted in zone 3. Paradoxically, clonal maintenance was anti-correlated with cancer initiation, as zone 3 mutant livers showed increased tumorigenesis. To define mechanisms, we individually deleted eleven zone specific genes in HCC mouse models, revealing that Gstm2 and Gstm3 were required for efficient HCC initiation in zone 3. These data indicate that liver cancer initiation is dependent on zonation but independent of clonal expansion.

cancer biology↗

bioRxiv2024

Target-aware Molecule Generation for Drug Design Using a Chemical Language Model

Generative drug design facilitates the creation of compounds effective against pathogenic target proteins. This opens up the potential to discover novel compounds within the vast chemical space and fosters the development of innovative therapeutic strategies. However, the practicality of generated molecules is often limited, as many designs focus on a narrow set of drug-related properties, failing to improve the success rate of subsequent drug discovery process. To overcome these challenges, we develop TamGen, a method that employs a GPT-like chemical language model and enables target-aware molecule generation and compound refinement. We demonstrate that the compounds generated by TamGen have improved molecular quality and viability. Additionally, we have integrated TamGen into a drug discovery pipeline and identified 7 compounds showing compelling inhibitory activity against the Tuberculosis ClpP protease, with the most effective compound exhibiting a half maximal inhibitory concentration (IC50) of 1.9 M. Our findings underscore the practical potential and real-world applicability of generative drug design approaches, paving the way for future advancements in the field.

biochemistry↗

bioRxiv2023

Benchmarking Reverse Docking through AlphaFold2 Human Proteome

Predicting binding of a small molecule to the human proteome by reverse docking methods, we can predict the target interactions of drug compounds in the human body, as well as further evaluate their potential off-target effects or toxic side effects. In this study, we constructed 11 pipelines to evaluate and benchmark thoroughly the predictive capabilities of these reverse docking pipelines. The pipelines were built using site prediction tools (PointSite and SiteMap) based on the AF2 human proteome, docking programs (Glide and AutoDock Vina), and scoring functions (Glide, Autodock Vina, RTMScore, DeepRMSD, OnionNet-SFCT). The results show that pipeline glide_sfct (PS) exhibited the best target prediction ability and successfully predicted the similar proteins of native targets. This finding provides important clues for understanding the promiscuity between the drug ligand and the whole human proteome. In general, our study has the potential to increase the success rate and reduce the development timeline of drug discovery, thereby saving costs.

bioinformatics↗

bioRxiv2023

ST-GEARS: Advancing 3D Downstream Research through Accurate Spatial Information Recovery

Three-dimensional Spatial Transcriptomics has revolutionized our understanding of tissue regionalization, organogenesis, and development. However, to reconstruct single sections back to their in situ three-dimensional morphology, existing approaches either only adopt gene expression information to guide reconstruction or overlook shape correction against experiment-induced section distortions. This leads to significant discrepancies between reconstruction results and the actual in vivo locations of cells, imposing unreliable spatial profiles to downstream analysis. To address these challenges, we propose ST-GEARS (Spatial Transcriptomics GEospatial profile recovery system through AnchoRS), which solves optimized anchors between in situ closest spots utilizing expression and structural similarity across sections and recovers in vivo spatial information under the guidance of anchors. By employing innovative Distributive Constraints into the Optimization scheme, it retrieves anchors with higher precision compared to existing methods. Taking these anchors as reference points, ST-GEARS first rigidly aligns sections, then introduces and infers Elastic Fields to counteract distortions. ST-GEARS denoises the fields using context information by Gaussian Denoising. Utilizing the denoised fields, it eliminates distortions and eventually recovers original spatial profile through innovative and mathematically proved Bi-sectional Fields Application. Studying ST-GEARS on both bi-sectional registration and complete tissue reconstruction across sectional distances and sequencing platforms, we observed its outstanding performance in spatial information recovery across tissue, cell, and gene levels compared to current approaches. Through this recovery, ST-GEARS provides precise and well-explainable gears between in vivo situations and 3D in vitro analysis, powerfully fueling the potential of biological discoveries.

bioinformatics↗

bioRxiv2023

Post-GWAS Prioritization of Genome-Phenome Associations in Sorghum

Genome-Wide Association Studies (GWAS) are widely used to infer the genetic basis of traits in organisms, yet selecting appropriate thresholds for analysis remains a significant challenge. In this study, we developed the Sequential SNP Prioritization Algorithm (SSPA) to elucidate the genetic underpinnings of two key phenotypes in Sorghum bicolor: maximum canopy height and maximum growth rate. Utilizing a subset of the Sorghum Bioenergy Association Panel cultivated at the Maricopa Agricultural Center in Arizona, our objective was to employ GWAS with specific permissive-filtered thresholds to identify the genetic markers associated with these traits, allowing for a broader collection of explanatory candidate genes. Following this, our proposed method incorporates a feature engineering approach based on statistical correlation coefficient to reveal patterns between phenotypic similarity and genetic proximity across 274 accessions. This approach helps prioritize Single Nucleotide Polymorphisms (SNPs) likely to be associated with the studied phenotype. Additionally, we evaluated the impact of SSPA by considering all variants (SNPs) as inputs, without any GWAS filtering, as a complementary analysis. Empirical evidence including ontology- based gene function, spatial and temporal expression, and similarity to known homologs, demonstrated that SSPA effectively prioritizes SNPs and genes influencing the phenotype of interest, providing valuable insights for functional genetics research.

genomics↗

bioRxiv2023

Effects of greening-induced warming and cooling on tree phenology in temperate and boreal forests

Tree phenology, periodic biological events in trees, is highly sensitive to climate change. It has been reported that forest greening can influence the local climate by altering the seasonal surface energy budget. However, tree phenological responses to forest greening remains poorly understood at large spatial scales. Combining remote-sensing derived phenological and leaf area indices since 2001, herein we show that forest greening led to earlier spring (-1.05 {+/-} 0.17 d) and autumn phenology (-1.95 {+/-} 0.14 d) in temperate and boreal forests. Our results show that forest greening in winter and spring decreased surface albedo and thus resulted in biophysical warming that caused earlier spring leaf phenology. In contrast, forest greening in summer and autumn triggered biophysical cooling by enhancing evapotranspiration, which led to earlier autumn leaf phenology. These findings suggest that forest greening could significantly alter tree phenology through seasonal biophysical impacts. Therefore, it is essential to incorporate these complicated biophysical impacts of greening into tree phenology models to accurately predict future shifts in tree phenology under future climate change.

ecology↗

bioRxiv2023

ThermoLink: Bridging Disulfide Bond and EnzymeThermostability through Database Construction andMachine Learning Prediction

Disulfide bonds (SS bonds), covalently formed by sulfur atoms in cysteine residues, play a crucial role in protein folding and structure stability. Due to their significance, artificial disulfide bonds are often introduced to enhance the thermostability of proteins. Although an increasing number of tools can assist with this task, significant amounts of time and resources were often wasted due to inadequate consideration. To enhance the accuracy and efficiency of designing disulfide bonds for protein thermostability improvement, we first collected disulfide-bond data with protein thermostability data from extensive literature sources. Then, various sequence- and structure-based features were extracted, and machine learning models were constructed to predict whether a disulfide bond could improve protein thermostability. Among all models, the neighborhood context model using the Adaboost-DT algorithm performs the best, and the AUC-ROC score and accuracy are 0.773 and 0.714, respectively. Alongside this, we also found that AlphaFold2 exhibits a high superiority in predicting disulfide bonds, and the coevolutionary relationship between residue pairs, to some extent, could also guide artificial disulfide-bond design. The SS-bond data has been integrated into an online server, named ThermoLink, available at guolab.mpu.edu.mo/thermoLink. O_TEXTBOXKey MessagesO_LIWe manually curated a database of disulfide bonds and their impacts on protein thermostability from the literature. C_LIO_LIAlphaFold2 exhibits a high superiority in predicting disulfide bonds, and to some extent, the conservation and co-evolutionary information of the residue pairs involved in disulfide bonds could also guide artificial disulfide-bond design. C_LIO_LIMachine learning models were developed to predict potential disulfide bonds with the aim of improving protein thermostability, offering an extended design strategy beginning with existing SS-bond prediction tools to optimize variant selection. C_LIO_LIThis work provides valuable data and theoretical support for a more precise design of disulfide bonds for improving protein thermostability. C_LI C_TEXTBOX

bioinformatics↗

bioRxiv2023

Binary Discriminator Facilitates GPT-based Protein Design

Generative pre-trained transformers (GPT) models provide powerful tools for de novo protein design (DNPD). GPT-based DNPD involves three procedures: a) finetuning the model with proteins of interest; b) generating sequence candidates with the finetuned model; and c) prioritizing the sequence candidates. Existing prioritization strategies heavily rely on sequence identity, undermining the diversity. Here, we coupled a protein GPT model with a custom discriminator, which enables selecting candidates of low identity to natural sequences while highly likely with desired functions. We applied this framework to creating novel antimicrobial peptides (AMPs) and malate dehydrogenases (MDHs). Experimental verification pinpointed four broad-spectrum AMPs from 24 candidates. Comprehensive computational analyses on the prioritized MDHs candidates provide compelling evidence for the anticipated function. During experimental validation, 4/10 and 3/10 natural MDHs and generated-prioritized novel candidates, respectively, were expressed and soluble. All the soluble candidates (3/3) are functional in vitro. This framework is time- and data-efficient and may therefore considerably expedite the DNPD process.

bioinformatics↗

bioRxiv2023

A New Paradigm for Applying Deep Learning to Protein-Ligand Interaction Prediction

Protein-ligand interaction prediction poses a significant challenge in the field of drug design. Numerous machine learning and deep learning models have been developed to identify the most accurate docking poses of ligands and active compounds against specific targets. However, the current models often suffer from inadequate accuracy and lack practical physical significance in their scoring systems. In this research paper, we introduce IGModel, a novel approach that leverages the geometric information of protein-ligand complexes as input for predicting the root mean square deviation (RMSD) of docking poses and the binding strength (the negative value of the logrithm of binding affinity, pKd) with the same prediction framework. By incorporating the geometric information, IGModel ensures that its scores carry intuitive meaning. The performance of IGModel has been extensively evaluated on various docking power test sets, including the CASF-2016 benchmark, PDBbind-CrossDocked-Core, and DISCO set, consistently achieving state-of-theart accuracies. Furthermore, we assess IGModels generalization ability and robustness by evaluating it on unbiased test sets and sets containing target structures generated by AlphaFold2. The exceptional performance of IGModel on these sets demonstrates its efficacy. Additionally, we visualize the latent space of protein-ligand interactions encoded by IGModel and conduct interpretability analysis, providing valuable insights. This study presents a novel framework for deep learning-based prediction of protein-ligand interactions, contributing to the advancement of this field. O_TEXTBOXKey MessagesO_LIWe introduce the first framework for simultaneously predicting the RMSD of the ligand docking pose and its binding strength to the target. C_LIO_LIIGModel can effectively improve the accuracy of identifying the near-native binding poses of the ligands, and can still outperform most baseline models in scoring power, ranking power and screening power tasks. C_LIO_LIIGModel is still ahead of other state-of-the-art models in the unbiased data set and the target structure predicted by AlphaFold2, proving its excellent generalization ability. C_LIO_LILatent space provided by IGModel learns the physical interactions, thus indicating the robustness of the model. C_LI C_TEXTBOX

bioinformatics↗