Tong Wang

Microsoft Research Asia (China)

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Tie‐Yan Liu

Microsoft Research Asia (China)

Bin Shao

Microsoft Research Asia (China)

Xinheng He

University of Chinese Academy of Sciences

Haipeng Gong

Tsinghua University

Mingyu Li

Wuhan University of Technology

Nanning Zheng

Xi'an Jiaotong University

Xin Zheng

Chinese Academy of Medical Sciences & Peking Union Medical College

Yusong Wang

Shanghai East Hospital

Yusong Wang

Anhui Agricultural University

Linqi Zhang

Peking University

Cooperative Institutions

Chinese Academy of Sciences

Tsinghua University

Huazhong University of Science and Technology

Microsoft Research Asia (China)

Chinese Academy of Medical Sciences & Peking Union Medical College

University of Chinese Academy of Sciences

Peking University

Shanghai Jiao Tong University

Xi'an Jiaotong University

Sichuan University

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

DSN-DDI: an accurate and generalized framework for drug–drug interaction prediction by dual-view representation learning

Briefings in Bioinformatics (2022)

Zimeng Li Shichao Zhu Bin Shao Xiangxiang Zeng Tong Wang

Drug-drug interaction (DDI) prediction identifies interactions of drug combinations in which the adverse side effects caused by the physicochemical incompatibility have attracted much attention. Previous studies usually model drug information from single or dual views of the whole drug molecules but ignore the detailed interactions among atoms, which leads to incomplete and noisy information and limits the accuracy of DDI prediction. In this work, we propose a novel dual-view drug representation learning network for DDI prediction ('DSN-DDI'), which employs local and global representation learning modules iteratively and learns drug substructures from the single drug ('intra-view') and the drug pair ('inter-view') simultaneously. Comprehensive evaluations demonstrate that DSN-DDI significantly improved performance on DDI prediction for the existing drugs by achieving a relatively improved accuracy of 13.01% and an over 99% accuracy under the transductive setting. More importantly, DSN-DDI achieves a relatively improved accuracy of 7.07% to unseen drugs and shows the usefulness for real-world DDI applications. Finally, DSN-DDI exhibits good transferability on synergistic drug combination prediction and thus can serve as a generalized framework in the drug discovery field.

Representation

Drug-drug interaction

Drug target

Transferability

10.1093/bib/bbac597

Cite

Citations (60)

A neural network with encoded visible edge prior for limited‐angle computed tomography reconstruction

Medical Physics (2021)

Genwei Ma Yinghui Zhang Xing Zhao Tong Wang Hongwei Li

Limited-angle computed tomography is a challenging but important task in certain medical and industrial applications for nondestructive testing. The limited-angle reconstruction problem is highly ill-posed and conventional reconstruction algorithms would introduce heavy artifacts. Various models and methods have been proposed to improve the quality of reconstructions by introducing different priors regarding to the projection data or ideal images. However, the assumed priors might not be practically applicable to all limited-angle reconstruction problems. Convolutional neural network (CNN) exhibits great promise in the modeling of data coupling and has recently become an important technique in medical imaging applications. Although existing CNN methods have demonstrated promising results, their robustness is still a concern. In this paper, in light of the theory of visible and invisible boundaries, we propose an alternating edge-preserving diffusion and smoothing neural network (AEDSNN) for limited-angle reconstruction that builds the visible boundaries as priors into its structure. The proposed method generalizes the alternating edge-preserving diffusion and smoothing (AEDS) method for limited-angle reconstruction developed in the literature by replacing its regularization terms by CNNs, by which the piecewise constant assumption assumed by AEDS is effectively relaxed.The AEDSNN is derived by unrolling the AEDS algorithm. AEDSNN consists of several blocks, and each block corresponds to one iteration of the AEDS algorithm. In each iteration of the AEDS algorithm, three subproblems are sequentially solved. So, each block of AEDSNN possesses three main layers: data matching layer, x -direction regularization layer for visible edges diffusion, and y -direction regularization layer for artifacts suppressing. The data matching layer is implemented by conventional ordered-subset simultaneous algebraic reconstruction technique (OS-SART) reconstruction algorithm, while the two regularization layers are modeled by CNNs for more intelligent and better encoding of priors regarding to the reconstructed images. To further strength the visible edge prior, the attention mechanism and the pooling layers are incorporated into AEDSNN to facilitate the procedure of edge-preserving diffusion from visible edges.We have evaluated the performance of AEDSNN by comparing it with popular algorithms for limited-angle reconstruction. Experiments on the medical dataset show that the proposed AEDSNN effectively breaks through the piecewise constant assumption usually assumed by conventional reconstruction algorithms, and works much better for piecewise smooth images with nonsharp edges. Experiments on the printed circuit board (PCB) dataset show that AEDSNN can better encode and utilize the visible edge prior, and its reconstructions are consistently better compared to the competing algorithms.A deep-learning approach for limited-angle reconstruction is proposed in this paper, which significantly outperforms existing methods. The superiority of AEDSNN consists of three aspects. First, by the virtue of CNN, AEDSNN is free of parameter-tuning. This is a great facility compared to conventional reconstruction methods; Second, AEDSNN is quite fast. Conventional reconstruction methods usually need hundreds even thousands of iterations, while AEDSNN just needs three to five iterations (i.e., blocks); Third, the learned regularizer by AEDSNN enjoys a broader application capacity, which could work well with piecewise smooth images and surpass the piecewise constant assumption frequently assumed for computed tomography images.

Smoothing

Regularization

10.1002/mp.15205

Cite

Citations (7)

Stochastic Lag Time Parameterization for Markov State Models of Protein Dynamics

The Journal of Physical Chemistry B (2022)

Shiqi Gong Xinheng He Qi Meng Zhi-Ming Ma Bin Shao

Markov state models (MSMs) play a key role in studying protein conformational dynamics. A sliding count window with a fixed lag time is widely used to sample sub-trajectories for transition counting and MSM construction. However, sub-trajectories sampled with a fixed lag time may not perform well under different selections of lag time, which requires strong prior practice and leads to less robust estimation. To alleviate it, we propose a novel stochastic method from a Poisson process to generate perturbative lag time for sub-trajectory sampling and utilize it to construct a Markov chain. Comprehensive evaluations on the double-well system, WW domain, BPTI, and RBD–ACE2 complex of SARS-CoV-2 reveal that our algorithm significantly increases the robustness and power of a constructed MSM without disturbing the Markovian properties. Furthermore, the superiority of our algorithm is amplified for slow dynamic modes in complex biological processes.

Robustness

Lag time

Sliding window protocol

10.1021/acs.jpcb.2c03711

Cite

Citations (3)

Prediction of Protein Quaternary Structure by a Novel Manifold Learning Algorithm

Tong Wang Jian Wang Lixiu Yao

With the explosion of protein sequences generated in the Post-Genomic Age, it is urgent to develop an automated method to predict protein quaternary structure. To explore this problem, we adopted an approach based on a sequence encoding descriptor by fusing PseAA (Pseudo Amino Acid) and DC (Dipeptide Composition) representing a protein sample. Here, a completely different approach, manifold learning algorithm MVP (Maximum variance projection) is introduced to extract the key features from the high-dimensional feature space. The dimension-reduced descriptor vector thus obtained is a compact representation of the original high dimensional vector. Our jackknife test results indicate that it is very promising to use the dimensionality reduction approaches to cope with complicated problems in biological systems, such as predicting the quaternary structure of proteins.

Feature vector

Representation

10.1109/isa.2011.5873419

Cite

Citations (0)

Immune escape mechanisms of severe fever with thrombocytopenia syndrome virus

Frontiers in Immunology (2022)

Tong Wang Ling Xu Bin Zhu Junzhong Wang Xin Zheng

Severe fever with thrombocytopenia syndrome (SFTS), which is caused by SFTS virus (SFTSV), poses a serious threat to global public health, with high fatalities and an increasing prevalence. As effective therapies and prevention strategies are limited, there is an urgent need to elucidate the pathogenesis of SFTS. SFTSV has evolved several mechanisms to escape from host immunity. In this review, we summarize the mechanisms through which SFTSV escapes host immune responses, including the inhibition of innate immunity and evasion of adaptive immunity. Understanding the pathogenesis of SFTS will aid in the development of new strategies for the treatment of this disease.

Pathogenesis

Evasion (ethics)

10.3389/fimmu.2022.937684

Cite

Citations (13)

Overcoming the barrier of orbital-free density functional theory for molecular systems using deep learning

Nature Computational Science (2024)

He Zhang Siyuan Liu Jiacheng You Chang Liu Shuxin Zheng

Representation

Quantum Chemistry

Basis (linear algebra)

10.1038/s43588-024-00605-8

Cite

Citations (8)

Identification of residue pairing in interacting β-strands from a predicted residue contact map

bioRxiv (Cold Spring Harbor Laboratory) (2017)

Wenzhi Mao Tong Wang Wenxuan Zhang Haipeng Gong

Abstract Despite the rapid progress of protein residue contact prediction, predicted residue contact maps frequently contain many errors. However, information of residue pairing in β strands could be extracted from a noisy contact map, due to the presence of characteristic contact patterns in β-β interactions. This information may benefit the tertiary structure prediction of mainly β proteins. In this work, we introduce a novel ridge-detection-based β-β contact predictor, RDb 2 C, to identify residue pairing in β strands from any predicted residue contact map. The algorithm adopts ridge detection, a well-developed technique in computer image processing, to capture consecutive residue contacts, and then utilizes a novel multi-stage random forest framework to integrate the ridge information and additional features for prediction. Starting from the predicted contact map of CCMpred, RDb 2 C remarkably outperforms all state-of-the-art methods on two conventional test sets of β proteins (BetaSheet916 and BetaSheet1452), and achieves F1-scores of ~62% and ~76% at the residue level and strand level, respectively. Taking the prediction of the more advanced RaptorX-Contact as input, RDb 2 C achieves impressively higher performance, with F1-scores reaching ~76% and ~86% at the residue level and strand level, respectively. According to our tests on 61 mainly β proteins, improvement in the β-β contact prediction can further ameliorate the structural prediction. Availability: All source data and codes are available at http://166.111.152.91/Downloads.html or at the GitHub address of https://github.com/wzmao/RDb2C . Author summary Due to the topological complexity, mainly β proteins are challenging targets in protein structure prediction. Knowledge of the pairing between β strands, especially the residue pairing pattern, can greatly facilitate the tertiary structure prediction of mainly β proteins. In this work, we developed a novel algorithm to identify the residue pairing in β strands from a predicted residue contact map. This method adopts the ridge detection technique to capture the characteristic pattern of β-β interactions from the map and then utilizes a multi-stage random forest framework to predict β-β contacts at the residue level. According to our tests, our method could effectively improve the prediction of β-β contacts even from a highly noisy contact map. Moreover, the refined β-β contact information could effectively improve the structural modeling of mainly β proteins.

Residue (chemistry)

10.1101/214643

Cite

Citations (0)

AIMD-Chig: Exploring the conformational space of a 166-atom protein Chignolin with ab initio molecular dynamics

Scientific Data (2023)

Tong Wang Xinheng He Mingyu Li Bin Shao Tie‐Yan Liu

Abstract Molecular dynamics (MD) simulations have revolutionized the modeling of biomolecular conformations and provided unprecedented insight into molecular interactions. Due to the prohibitive computational overheads of ab initio simulation for large biomolecules, dynamic modeling for proteins is generally constrained on force field with molecular mechanics, which suffers from low accuracy as well as ignores the electronic effects. Here, we report AIMD-Chig, an MD dataset including 2 million conformations of 166-atom protein Chignolin sampled at the density functional theory (DFT) level with 7,763,146 CPU hours. 10,000 conformations were initialized covering the whole conformational space of Chignolin , including folded, unfolded, and metastable states. Ab initio simulations were driven by M06-2X/6-31 G* with a Berendsen thermostat at 340 K. We reported coordinates, energies, and forces for each conformation. AIMD-Chig brings the DFT level conformational space exploration from small organic molecules to real-world proteins. It can serve as the benchmark for developing machine learning potentials for proteins and facilitate the exploration of protein dynamics with ab initio accuracy.

Force Field

Metastability

SIESTA (computer program)

10.1038/s41597-023-02465-9

Cite

Citations (11)

M-OFDFT: Overcoming the Barrier of Orbital-Free Density Functional Theory for Molecular Systems Using Deep Learning

arXiv (Cornell University) (2023)

He Zhang Siyuan Liu Jiacheng You Chang Liu Shuxin Zheng

Orbital-free density functional theory (OFDFT) is a quantum chemistry formulation that has a lower cost scaling than the prevailing Kohn-Sham DFT, which is increasingly desired for contemporary molecular research. However, its accuracy is limited by the kinetic energy density functional, which is notoriously hard to approximate for non-periodic molecular systems. Here we propose M-OFDFT, an OFDFT approach capable of solving molecular systems using a deep learning functional model. We build the essential non-locality into the model, which is made affordable by the concise density representation as expansion coefficients under an atomic basis. With techniques to address unconventional learning challenges therein, M-OFDFT achieves a comparable accuracy with Kohn-Sham DFT on a wide range of molecules untouched by OFDFT before. More attractively, M-OFDFT extrapolates well to molecules much larger than those seen in training, which unleashes the appealing scaling of OFDFT for studying large molecules including proteins, representing an advancement of the accuracy-efficiency trade-off frontier in quantum chemistry.

Basis (linear algebra)

Representation

Quantum Chemistry

10.48550/arxiv.2309.16578

Cite

Citations (0)

MIHIC: A multiplex IHC histopathological image classification dataset for lung cancer immune microenvironment quantification

Zenodo (CERN European Organization for Nuclear Research) (2023)

Ranran Wang Yusong Qiu Tong Wang Mingkang Wang Shan Jin

A cohort of 47 TMA sections from 114 patients was collected from Liaoning cancer hospital \& Institute, where each TMA section has the size of 188,416$\times$110,080 pixels (i.e., 42660.87um$\times$24924.15um) at 40$\times$ magnification. TMA sections contain different number of tissue cores, ranging from 28 to 48. After excluding poor quality TMA sections with tissue folding, missing or contamination, there are totally 114 patients. Each patient has tissue cores with 12 different IHC stains, including CD3, CD20, CD34, CD38, CD68, CDK4, cyclin-D1, D2-40, FAP, Ki67, P53, and SMA. Two pathologists have manually labeled clear tissue regions (i.e., without controversy) in TMA sections based on visual examination via Qupath software, where six tissue types including Alveoli, Immune cells, Nerosis, Other, Stroma, Tumor were annotated. Besides the annotated six tissue types, we added one more Background type. To build histological classification models, we split 309,698 image patches in MIHIC dataset into three sets: training, validation and test. Note that image patches extracted from the same annotated tissue region are distributed into the same set, which avoids data leakage during classification model optimization. According to the number of extracted ROIs, train, val and test accounted for 64\%, 16\% and 20\%.

Multiplex

10.5281/zenodo.10065509

Cite

Citations (0)