JZJianyang Zeng

Jianyang Zeng

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Jianyang Zeng has a total of 58 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Crowdsourced mapping extends the target space of kinase inhibitors

    by Anna Cichonska, Balaguru Ravikumar, Robert J Allaway, Sungjoon Park, Fangping Wan, Olexandr Isayev, Shuya Li, Michael J Mason, Andrew Lamb, Zia-Ur-Rehman Tanoli, et al.

    Despite decades of intensive search for compounds that modulate the activity of particular targets, there are currently small-molecules available only for a small proportion of the human proteome. Effective approaches are therefore required to map the massive space of unexplored compound-target interactions for novel and potent activities. Here, we carried out a crowdsourced benchmarking of predictive models for kinase inhibitor potencies across multiple kinase families using unpublished bioactivity data. The top-performing predictions were based on kernel learning, gradient boosting and deep learning, and their ensemble resulted in predictive accuracy exceeding that of kinase activity assays. We then made new experiments based on the model predictions, which further improved the accuracy of experimental mapping efforts and identified unexpected potencies even for under-studied kinases. The open-source algorithms together with the novel bioactivities between 95 compounds and 295 kinases provide a resource for benchmarking new prediction algorithms and for extending the druggable kinome.


  • Open Access

    A deep boosting based approach for capturing the sequence binding preferences of RNA-binding proteins from high-throughput CLIP-seq data

    by Shuya Li, Fanghong Dong, Yuexin Wu, Sai Zhang, Chen Zhang, Xiao Liu, Tao Jiang, Jianyang Zeng

    Characterizing the binding behaviors of RNA-binding proteins (RBPs) is important for understanding their functional roles in gene expression regulation. However, current high-throughput experimental methods for identifying RBP targets, such as CLIP-seq and RNAcompete, usually suffer from the false negative issue. Here, we develop a deep boosting based machine learning approach, called DeBooster, to accurately model the binding sequence preferences and identify the corresponding binding targets of RBPs from CLIP-seq data. Comprehensive validation tests have shown that DeBooster can outperform other state-of-the-art approaches in RBP target prediction. In addition, we have demonstrated that DeBooster may provide new insights into understanding the regulatory functions of RBPs, including the binding effects of the RNA helicase MOV10 on mRNA degradation, the potentially different ADAR1 binding behaviors related to its editing activity, as well as the antagonizing effect of RBP binding on miRNA repression. Moreover, DeBooster may provide an effective index to investigate the effect of pathogenic mutations in RBP binding sites, especially those related to splicing events. We expect that DeBooster will be widely applied to analyze large-scale CLIP-seq experimental data and can provide a practically useful tool for novel biological discoveries in understanding the regulatory mechanisms of RBPs. The source code of DeBooster can be downloaded from http://github.com/dongfanghong/deepboost.


  • Open Access

    Metagenomic binning through low density hashing

    by Yunan Luo, Y. William Yu, Jianyang Zeng, Bonnie Berger, Jian Peng

    Bacterial microbiomes of incredible complexity are found throughout the world, from exotic marine locations to the soil in our yards to within our very guts. With recent advances in Next-Generation Sequencing (NGS) technologies, we have vastly greater quantities of microbial genome data, but the nature of environmental samples is such that DNA from different species are mixed together. Here, we present Opal for metagenomic binning, the task of identifying the origin species of DNA sequencing reads. Our Opal method introduces low-density, even-coverage hashing to bioinformatics applications, enabling quick and accurate metagenomic binning. Our tool is up to two orders of magnitude faster than leading alignment-based methods at similar or improved accuracy, allowing computational tractability on large metagenomic datasets. Moreover, on public benchmarks, Opal is substantially more accurate than both alignment-based and alignment-free methods (e.g. on SimHC20.500, Opal achieves 95% F1-score while Kraken and CLARK achieve just 91% and 88%, respectively); this improvement is likely due to the fact that the latter methods cannot handle computationally-costly long-range dependencies, which our even-coverage, low-density fingerprints resolve. Notably, capturing these long-range dependencies drastically improves Opal’s ability to detect unknown species that share a genus or phylum with known bacteria. Additionally, the family of hash functions Opal uses can be generalized to other sequence analysis tasks that rely on k-mer based methods to encode long-range dependencies.


  • Open Access

    TITER: predicting translation initiation sites by deep learning

    by Sai Zhang, Hailin Hu, Tao Jiang, Lei Zhang, Jianyang Zeng

    Motivation: Translation initiation is a key step in the regulation of gene expression. In addition to the annotated translation initiation sites (TISs), the translation process may also start at multiple alternative TISs (including both AUG and non-AUG codons), which makes it challenging to predict TISs and study the underlying regulatory mechanisms. Meanwhile, the advent of several high-throughput sequencing techniques for profiling initiating ribosomes at single-nucleotide resolution, e.g., GTI-seq and QTI-seq, provides abundant data for systematically studying the general principles of translation initiation and the development of computational method for TIS identification. Methods: We have developed a deep learning based framework, named TITER, for accurately predicting TISs on a genome-wide scale based on QTI-seq data. TITER extracts the sequence features of translation initiation from the surrounding sequence contexts of TISs using a hybrid neural network and further integrates the prior preference of TIS codon composition into a unified prediction framework. Results: Extensive tests demonstrated that TITER can greatly outperform the state-of-the-art prediction methods in identifying TISs. In addition, TITER was able to identify important sequence signatures for individual types of TIS codons, including a Kozak-sequence-like motif for AUG start codon. Furthermore, the TITER prediction score can be related to the strength of translation initiation in various biological scenarios, including the repressive effect of the upstream open reading frames (uORFs) on gene expression and the mutational effects influencing translation initiation efficiency. Availability: TITER is available as an open-source software and can be downloaded from https://github.com/zhangsaithu/titer Contact: lzhang20@mail.tsinghua.edu.cn and zengjy321@tsinghua.edu.cn


  • Open Access

    A Network Integration Approach for Drug-Target Interaction Prediction and Computational Drug Repositioning from Heterogeneous Information

    by Yunan Luo, Xinbin Zhao, Jingtian Zhou, Jinglin Yang, Yanqing Zhang, Wenhua Kuang, Jian Peng, Ligong Chen, Jianyang Zeng

    The emergence of large-scale genomic, chemical and pharmacological data provides new opportunities for drug discovery and repositioning. Systematic integration of these heterogeneous data not only serves as a promising tool for identifying new drug-target interactions (DTIs), which is an important step in drug development, but also provides a more complete understanding of the molecular mechanisms of drug action. In this work, we integrate diverse drug-related information, including drugs, proteins, diseases and side-effects, together with their interactions, associations or similarities, to construct a heterogeneous network with 12,015 nodes and 1,895,445 edges. We then develop a new computational pipeline, called DTINet, to predict novel drug-target interactions from the constructed heterogeneous network. Specifically, DTINet focuses on learning a low-dimensional vector representation of features for each node, which accurately explains the topological properties of individual nodes in the heterogeneous network, and then predicts the likelihood of a new DTI based on these representations via a vector space projection scheme. DTINet achieves substantial performance improvement over other state-of-the-art methods for DTI prediction. Moreover, we have experimentally validated the novel interactions between three drugs and the cyclooxygenase (COX) protein family predicted by DTINet, and demonstrated the new potential applications of these identified COX inhibitors in preventing inflammatory diseases. These results indicate that DTINet can provide a practically useful tool for integrating heterogeneous information to predict new drug-target interactions and repurpose existing drugs. The source code of DTINet and the input heterogeneous network data can be downloaded from http://github.com/luoyunan/DTINet .


  • Open Access

    Deep learning with feature embedding for compound-protein interaction prediction

    by Fangping Wan, Jianyang (Michael) Zeng

    Accurately identifying compound-protein interactions in silico can deepen our understanding of the mechanisms of drug action and significantly facilitate the drug discovery and development process. Traditional similarity-based computational models for compound-protein interaction prediction rarely exploit the latent features from current available large-scale unlabelled compound and protein data, and often limit their usage on relatively small-scale datasets. We propose a new scheme that combines feature embedding (a technique of representation learning) with deep learning for predicting compound-protein interactions. Our method automatically learns the low-dimensional implicit but expressive features for compounds and proteins from the massive amount of unlabelled data. Combining effective feature embedding with powerful deep learning techniques, our method provides a general computational pipeline for accurate compound-protein interaction prediction, even when the interaction knowledge of compounds and proteins is entirely unknown. Evaluations on current large-scale databases of the measured compound-protein affinities, such as ChEMBL and BindingDB, as well as known drug-target interactions from DrugBank have demonstrated the superior prediction performance of our method, and suggested that it can offer a useful tool for drug development and drug repositioning.


  • cOSPREY: A Cloud-Based Distributed Algorithm for Large-Scale Computational Protein Design

    by Yuchao Pan, Yuxi Dong, Jingtian Zhou, Mark Hallen, Bruce R. Donald, Jianyang Zeng, Wei Xu

    Finding the global minimum energy conformation (GMEC) of a huge combinatorial search space is the key challenge in computational protein design (CPD) problems. Traditional algorithms lack a scalable and efficient distributed design scheme, preventing researchers from taking full advantage of current cloud infrastructures. We design cloud OSPREY (cOSPREY), an extension to a widely used protein design software OSPREY, to allow the original design framework to scale to the commercial cloud infrastructures. We propose several novel designs to integrate both algorithm and system optimizations, such as GMEC-specific pruning, state search partitioning, asynchronous algorithm state sharing, and fault tolerance. We evaluate cOSPREY on three different cloud platforms using different technologies and show that it can solve a number of large-scale protein design problems that have not been possible with previous approaches.