JWJiansheng Wu

Jiansheng Wu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Jiansheng Wu has a total of 76 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Improving the Hit Rates of Virtual Screening by Active Learning from Bioactivity Feedback

    by Xun Deng, Junlong Liu, Zhike Liu, Jiansheng Wu, Fuli Feng, Jieping Ye, Zheng Wang

    Virtual screening has been widely used to identify potential hit candidates that can bind to the target protein in drug discovery. Contemporary screening methods typically rely on oversimplified scoring functions, frequently yielding one-digit hit rates (or even zero) among top-ranked candidates. The substantial cost of laboratory validation further constrains the exploration of candidate molecules. We find that test-time prediction refinement is almost blank in this area, which means bioactivity feedback in the wet-lab experiments is neglected. Here, we introduce an Active Learning from Bioactivity Feedback (ALBF) framework to enhance the weak hit rate of current virtual screening methods. ALBF spends the budget of wet-lab experiments iteratively and leverages the target-specific bioactivity insights from current wet-lab tests to refine the score results (i.e., rankings). Our framework consists of two components: a novel query strategy that considers the evaluation quality and its overall influence on other top-scored molecules; and an efficient score optimization strategy that propagates the bioactivity feedback to structurally similar molecules. We evaluated ALBF on diverse subsets of the well-known DUD-E and LIT-PCBA benchmarks. Our active learning protocol averagely enhances top-100 hit rates by 60% and 30% on DUD-E and LIT-PCBA with 50 to 200 bioactivity queries on the selected molecules that are deployed in ten rounds. The consistently superior performance demonstrates ALBF's potential to enhance both the accuracy and cost-effectiveness of active learning-based laboratory testing.


  • Open Access

    A novel self-attentive deep learning for predicting pregnancy outcome after hysteroscopic adhesiolysis in patients with intrauterine adhesions

    by Yueqi Li, Jieying Zheng, Yong Shi, Jiansheng Wu, Danhong Peng, Jun Wang

    This study presents SDL-IUAs, a novel deep learning framework that employs a self-attention mechanism to predict pregnancy outcomes in patients with intrauterine adhesions (IUAs) following hysteroscopic adhesiolysis. By integrating random forest-based feature importance analysis with self-attention, the model achieves enhanced predictive accuracy and interpretability. The experiments were conducted on a clinical dataset of 121 patients with 163 features, where 16.05% missing values were effectively handled using a matrix decomposition approach. SDL-IUAs demonstrated superior performance compared to traditional models, attaining an AUC of 0.96, a precision of 0.95, a recall of 0.94, and an F1-score of 0.94. Feature importance analysis highlighted age as a key predictive factor, providing actionable insights for clinical decision-making and personalized treatment planning. These results underscore the potential of SDL-IUAs as a robust, interpretable, and clinically applicable tool for pregnancy outcome prediction in IUA patients. The methodological advancements introduced in this study are expected to catalyze further research and innovation in this domain, ultimately contributing to improved clinical outcomes for IUA patients.


  • Vina-GPU 2.1: Towards Further Optimizing Docking Speed and Precision of AutoDock Vina and Its Derivatives

    by Shidi Tang, Ji Ding, Xiangyu Zhu, Zheng Wang, Haitao Zhao, Jiansheng Wu

    AutoDock Vina and its derivatives have established themselves as a prevailing pipeline for virtual screening in contemporary drug discovery. Our Vina-GPU method leverages the parallel computing power of GPUs to accelerate AutoDock Vina, and Vina-GPU 2.0 further enhances the speed of AutoDock Vina and its derivatives. Given the prevalence of large virtual screens in modern drug discovery, the improvement of speed and accuracy in virtual screening has become a longstanding challenge. In this study, we propose Vina-GPU 2.1, aimed at enhancing the docking speed and precision of AutoDock Vina and its derivatives through the integration of novel algorithms to facilitate improved docking and virtual screening outcomes. Building upon the foundations laid by Vina-GPU 2.0, we introduce a novel algorithm, namely Reduced Iteration and Low Complexity BFGS (RILC-BFGS), designed to expedite the most time-consuming operation. Additionally, we implement grid cache optimization to further enhance the docking speed. Furthermore, we employ optimal strategies to individually optimize the structures of ligands, receptors, and binding pockets, thereby enhancing the docking precision. To assess the performance of Vina-GPU 2.1, we conduct extensive virtual screening experiments on three prominent targets, utilizing two fundamental compound libraries and seven docking tools. Our results demonstrate that Vina-GPU 2.1 achieves an average 4.97-fold acceleration in docking speed and an average 342% improvement in EF1% compared to Vina-GPU 2.0. The source code and tools for Vina-GPU 2.1 are freely available accompanied by comprehensive instructions and illustrative examples.


  • Open Access

    OLB-AC: toward optimizing ligand bioactivities through deep graph learning and activity cliffs

    by Yueming Yin, Haifeng Hu, Jitao Yang, Chun Ye, Wilson Wen Bin Goh, Adams Wai-Kin Kong, Jiansheng Wu

    Deep graph learning (DGL) has been widely employed in the realm of ligand-based virtual screening. Within this field, a key hurdle is the existence of activity cliffs (ACs), where minor chemical alterations can lead to significant changes in bioactivity. In response, several DGL models have been developed to enhance ligand bioactivity prediction in the presence of ACs. Yet, there remains a largely unexplored opportunity within ACs for optimizing ligand bioactivity, making it an area ripe for further investigation. We present a novel approach to simultaneously predict and optimize ligand bioactivities through DGL and ACs (OLB-AC). OLB-AC possesses the capability to optimize ligand molecules located near ACs, providing a direct reference for optimizing ligand bioactivities with the matching of original ligands. To accomplish this, a novel attentive graph reconstruction neural network and ligand optimization scheme are proposed. Attentive graph reconstruction neural network reconstructs original ligands and optimizes them through adversarial representations derived from their bioactivity prediction process. Experimental results on nine drug targets reveal that out of the 667 molecules generated through OLB-AC optimization on datasets comprising 974 low-activity, noninhibitor, or highly toxic ligands, 49 are recognized as known highly active, inhibitor, or nontoxic ligands beyond the datasets’ scope. The 27 out of 49 matched molecular pairs generated by OLB-AC reveal novel transformations not present in their training sets. The adversarial representations employed for ligand optimization originate from the gradients of bioactivity predictions. Therefore, we also assess OLB-AC’s prediction accuracy across 33 different bioactivity datasets. Results show that OLB-AC achieves the best Pearson correlation coefficient (r2) on 27/33 datasets, with an average improvement of 7.2%–22.9% against the state-of-the-art bioactivity prediction methods. The code and dataset developed in this work are available at github.com/Yueming-Yin/OLB-AC.


  • Open Access

    Vina-GPU 2.1: towards further optimizing docking speed and precision of AutoDock Vina and its derivatives

    by Shidi Tang, Ji Ding, Xiangyu Zhu, Zheng Wang, Haitao Zhao, Jiansheng Wu

    AutoDock Vina and its derivatives have established themselves as a prevailing pipeline for virtual screening in contemporary drug discovery. Our Vina-GPU method leverages the parallel computing power of GPUs to accelerate AutoDock Vina, and Vina-GPU 2.0 further enhances the speed of AutoDock Vina and its derivatives. Given the prevalence of large virtual screens in modern drug discovery, the improvement of speed and accuracy in virtual screening has become a longstanding challenge. In this study, we propose Vina-GPU 2.1, aimed at enhancing the docking speed and precision of AutoDock Vina and its derivatives through the integration of novel algorithms to facil-itate improved docking and virtual screening outcomes. Building upon the foundations laid by Vina-GPU 2.0, we introduce a novel algorithm, namely Reduced Iteration and Low Complexity BFGS (RILC-BFGS), designed to expedite the most time-consuming operation. Additionally, we implement grid cache optimization to further enhance the docking speed. Furthermore, we employ optimal strategies to individually optimize the structures of ligands, receptors, and binding pockets, thereby enhancing the docking precision. To assess the performance of Vina-GPU 2.1, we conduct extensive virtual screening experiments on three prominent targets, utilizing two fundamental compound libraries and seven docking tools. Our results demonstrate that Vina-GPU 2.1 achieves an average 4.97-fold acceleration in docking speed and an average 342% improvement in EF1% compared to Vina-GPU 2.0. The source code and tools for Vina-GPU 2.1 are freely available at https://github.com/DeltaGroupNJUPT/Vina-GPU-2.1 , accompanied by comprehensive instructions and illustrative examples.


  • Effectiveness Analysis of Multiple Initial States Simulated Annealing Algorithm, a Case Study on the Molecular Docking Tool AutoDock Vina

    by Xingxing Zhou, Ming Ling, Qingde Lin, Shidi Tang, Jiansheng Wu, Haifeng Hu

    Simulated Annealing (SA) algorithm is not effective with large optimization problems for its slow convergence. Hence, several parallel Simulated Annealing (pSA) methods have been proposed, where the increase of searching threads can boost the speed of convergence. Although satisfactory solutions can be obtained by these methods, there is no rigorous mathematical analyses on their effectiveness. Thus, this paper introduces a probabilistic model, on which a theorem about the effectiveness of multiple initial states parallel SA (MISPSA) has been proven. The theorem also demonstrates that the increasing parallelism in pSA algorithm with the reducing of search depth in each thread could obtain almost the same probability of finding the global optimal solution. We validated our theorem on AutoDock Vina, a widely used molecular docking tool with high accuracy and docking speed. AutoDock Vina uses a pSA strategy to find optimal molecular conformations. Under the premise that the total searching workload (i.e., thread number * iteration depth of each thread) remains unchanged, the docking accuracy from an aggressively parallelized SA searching method is almost the same or even better than those from the default exhaustiveness (parallelism degree) configuration of AutoDock Vina. Taking complex '1hnn' as an example,with the increase (125x) in the number of initial states (from 8 to 1000) and the decrease in the search depth for each thread (from 15540 to 124, or 1/125 of the original search depth), the mean energy is -7.80 and -7.94, while the mean RMSD is 3.4 and 3.14, respectively. The result also implies that a considerable speedup (in this case 125x in theory) can be obtained by a highly parallelized SA algorithm implementation.


  • Vina-GPU 2.0: Further Accelerating AutoDock Vina and Its Derivatives with Graphics Processing Units

    by Ji Ding, Shidi Tang, Zheming Mei, Lingyue Wang, Qinqin Huang, Haifeng Hu, Ming Ling, Jiansheng Wu

    Modern drug discovery typically faces large virtual screens from huge compound databases where multiple docking tools are involved for meeting various real scenes or improving the precision of virtual screens. Among these tools, AutoDock Vina and its numerous derivatives are the most popular and have become the standard pipeline for molecular docking in modern drug discovery. Our recent Vina-GPU method realized 14-fold acceleration against AutoDock Vina on a piece of NVIDIA RTX 3090 GPU in one virtual screening case. Further speedup of AutoDock Vina and its derivatives with graphics processing units (GPUs) is beneficial to systematically push their popularization in large-scale virtual screens due to their high benefit–cost ratio and easy operation for users. Thus, we proposed the Vina-GPU 2.0 method to further accelerate AutoDock Vina and the most common derivatives with new docking algorithms (QuickVina 2 and QuickVina-W) with GPUs. Caused by the discrepancy in their docking algorithms, our Vina-GPU 2.0 adopts different GPU acceleration strategies. In virtual screening for two hot protein kinase targets, RIPK1 and RIPK3, from the DrugBank database, our Vina-GPU 2.0 reaches an average of 65.6-fold, 1.4-fold, and 3.6-fold docking acceleration against the original AutoDock Vina, QuickVina 2, and QuickVina-W while ensuring their comparable docking accuracy. In addition, we develop a friendly and installation-free graphical user interface tool for their convenient usage. The codes and tools of Vina-GPU 2.0 are freely available at https://github.com/DeltaGroupNJUPT/Vina-GPU-2.0, coupled with explicit instructions and examples.


  • Open Access

    An open unified deep graph learning framework for discovering drug leads

    by Yueming Yin, Haifeng Hu, Zhen Yang, Jitao Yang, Chun Ye, Jiansheng Wu, Wilson Wen Bin Goh

    Computational discovery of ideal lead compounds is a critical process for modern drug discovery. It comprises multiple stages: hit screening, molecular property prediction, and molecule optimization. Current efforts are disparate, involving the establishment of models for each stage, followed by multi-stage multi-model integration. However, this is non-ideal, as clumsy integration of incompatible models increases research overheads, and may even reduce success rates in drug discovery. Facilitating compatibilities requires establishing inherent model consistencies across lead discovery stages. Towards that effect, we propose an open deep graph learning (DGL) based pipeline: generative adversarial feature subspace enhancement (GAFSE), which first unifies the modeling of these stages into one learning framework. GAFSE also offers standardized modular design and streamlined interfaces for future expansions and community support. GAFSE combines adversarial/generative learning, graph attention network, graph reconstruction network, and optimizes the classification/regression loss, adversarial/generative loss, and reconstruction loss simultaneously. Convergence analysis theoretically guarantees model generalization performance. Exhaustive benchmarking demonstrates that the GAFSE pipeline achieves excellent performance across almost all lead discovery stages, while also providing valuable model interpretability. Hence, we believe this tool will enhance the efficiency and productivity of drug discovery researchers.

  • Open Access

    Accelerating AutoDock Vina with GPUs

    by Shidi Tang, Ruiqi Chen, Mengru Lin, Qingde Lin, Yanxiang Zhu, Ji Ding, Haifeng Hu, Ming Ling, Jiansheng Wu

    AutoDock Vina is one of the most popular molecular docking tools. In the latest benchmark CASF-2016 for comparative assessment of scoring functions, AutoDock Vina won the best docking power among all the docking tools. Modern drug discovery is facing a common scenario of large virtual screening of drug hits from huge compound databases. Due to the seriality characteristic of the AutoDock Vina algorithm, there is no successful report on its parallel acceleration with GPUs. Current acceleration of AutoDock Vina typically relies on the stack of computing power as well as the allocation of resource and tasks, such as the VirtualFlow platform. The vast resource expenditure and the high access threshold of users will greatly limit the popularity of AutoDock Vina and the flexibility of its usage in modern drug discovery. In this work, we proposed a new method, Vina-GPU, for accelerating AutoDock Vina with GPUs, which is greatly needed for reducing the investment for large virtual screens and also for wider application in large-scale virtual screening on personal computers, station servers or cloud computing, etc. Our proposed method is based on a modified Monte Carlo using simulating annealing AI algorithm. It greatly raises the number of initial random conformations and reduces the search depth of each thread. Moreover, a classic optimizer named BFGS is adopted to optimize the ligand conformations during the docking progress, before a heterogeneous OpenCL implementation was developed to realize its parallel acceleration leveraging thousands of GPU cores. Large benchmark tests show that Vina-GPU reaches an average of 21-fold and a maximum of 50-fold docking acceleration against the original AutoDock Vina while ensuring their comparable docking accuracy, indicating its potential for pushing the popularization of AutoDock Vina in large virtual screens.


  • AFSE: towards improving model generalization of deep graph learning of ligand bioactivities targeting GPCR proteins

    by Yueming Yin, Haifeng Hu, Zhen Yang, Feihu Jiang, Yihe Huang, Jiansheng Wu

    Ligand molecules naturally constitute a graph structure. Recently, many excellent deep graph learning (DGL) methods have been proposed and used to model ligand bioactivities, which is critical for the virtual screening of drug hits from compound databases in interest. However, pharmacists can find that these well-trained DGL models usually are hard to achieve satisfying performance in real scenarios for virtual screening of drug candidates. The main challenges involve that the datasets for training models were small-sized and biased, and the inner active cliff cases would worsen model performance. These challenges would cause predictors to overfit the training data and have poor generalization in real virtual screening scenarios. Thus, we proposed a novel algorithm named adversarial feature subspace enhancement (AFSE). AFSE dynamically generates abundant representations in new feature subspace via bi-directional adversarial learning, and then minimizes the maximum loss of molecular divergence and bioactivity to ensure local smoothness of model outputs and significantly enhance the generalization of DGL models in predicting ligand bioactivities. Benchmark tests were implemented on seven state-of-the-art open-source DGL models with the potential of modeling ligand bioactivities, and precisely evaluated by multiple criteria. The results indicate that, on almost all 33 GPCRs datasets and seven DGL models, AFSE greatly improved their enhancement factor (top-10%, 20% and 30%), which is the most important evaluation in virtual screening of hits from compound databases, while ensuring the superior performance on RMSE and $r^2$. The web server of AFSE is freely available at http://noveldelta.com/AFSE for academic purposes.