Jian Tang

Biography

Jian Tang is an Associate professor at HEC's Department of Decision Sciences. He is also an Adjunct professor at the Department of Computer Science and Operations Research at University of Montreal and a Core Academic member at Mila - Quebec AI Institute. He is a Canada CIFAR AI Chair and the Founder of BioGeometry, an AI startup that focuses on generative AI for antibody discovery. Tang’s main research interests are deep generative models and graph machine learning, and their applications to drug discovery. He is an international leader in graph machine learning, and LINE, his node representation method, has been widely recognized and cited more than five thousand times. He has also done pioneering work on AI for drug discovery, such as developing the first open-source machine learning frameworks for drug discovery, TorchDrug and TorchProtein.

Current Students

Huiyu Cai

PhD - Université de Montréal

huiyu.cai@mila.quebec

Collaborating researcher

Farzaneh Heidari

PhD - Université de Montréal

Principal supervisor :

Guillaume Rabusseau

Jerry Lu

PhD - Université de Montréal

Shentong Mo

Collaborating researcher - Carnegie Mellon University

Chence Shi

PhD - Université de Montréal

Sophie Xhonneux

PhD - Université de Montréal

Principal supervisor :

Gauthier Gidel

lpxhonneux@gmail.com

Xinyu Yuan

PhD - Université de Montréal

xinyu.yuan@mila.quebec

Zhihao Zhan

PhD - Université de Montréal

Zuobai Zhang

PhD - Université de Montréal

zhaocheng.zhu@umontreal.ca

Jianan Zhao

PhD - Université de Montréal

PhD - Université de Montréal

Publications

Pre-training Protein Structure Encoder via Siamese Diffusion Trajectory Prediction

Zuobai Zhang

Minghao Xu

Aurelie Lozano

Vijil Chenthamarakshan

Payel Das

Due to the determining role of protein structures on diverse protein functions, pre-training representations of proteins on massive unlabele… (see more)d protein structures has attracted rising research interests. Among recent efforts on this direction, mutual information (MI) maximization based methods have gained the superiority on various downstream benchmark tasks. The core of these methods is to design correlated views that share common information about a protein. Previous view designs focus on capturing structural motif co-occurrence on the same protein structure, while they cannot capture detailed atom/residue interactions. To address this limitation, we propose the Siamese Diffusion Trajectory Prediction (SiamDiff) method. SiamDiff builds a view as the trajectory that gradually approaches protein native structure from scratch, which facilitates the modeling of atom/residue interactions underlying the protein structural dynamics. Specifically, we employ the multimodal diffusion process as a faithful simulation of the structure-sequence co-diffusion trajectory, where rich patterns of protein structural changes are embedded. On such basis, we design a principled theoretical framework to maximize the MI between correlated multimodal diffusion trajectories. We study the effectiveness of SiamDiff on both residue-level and atom-level structures. On the EC and ATOM3D benchmarks, we extensively compare our method with previous protein structure pre-training approaches. The experimental results verify the consistently superior or competitive performance of SiamDiff on all benchmark tasks compared to existing baselines. The source code will be made public upon acceptance.

2023-02-01

ICLR.cc/2023/Conference (rejected)

Protein Representation Learning by Geometric Structure Pretraining

Zuobai Zhang

Minghao Xu

Arian Rokkum Jamasb

Vijil Chenthamarakshan

Aurelie Lozano

Payel Das

Learning effective protein representations is critical in a variety of tasks in biology such as predicting protein function or structure. Ex… (see more)isting approaches usually pretrain protein language models on a large number of unlabeled amino acid sequences and then finetune the models with some labeled data in downstream tasks. Despite the effectiveness of sequence-based approaches, the power of pretraining on known protein structures, which are available in smaller numbers only, has not been explored for protein property prediction, though protein structures are known to be determinants of protein function. In this paper, we propose to pretrain protein representations according to their 3D structures. We first present a simple yet effective encoder to learn the geometric features of a protein. We pretrain the protein graph encoder by leveraging multiview contrastive learning and different self-prediction tasks. Experimental results on both function prediction and fold classification tasks show that our proposed pretraining methods outperform or are on par with the state-of-the-art sequence-based methods, while using much less pretraining data. Our implementation is available at https://github.com/DeepGraphLearning/GearNet.

2023-02-01

ICLR.cc/2023/Conference (poster)

Protein Sequence and Structure Co-Design with Equivariant Translation

Chence Shi

Chuanrui Wang

Jiarui Lu

Bozitao Zhong

Proteins are macromolecules that perform essential functions in all living organisms. Designing novel proteins with specific structures and … (see more)desired functions has been a long-standing challenge in the field of bioengineering. Existing approaches generate both protein sequence and structure using either autoregressive models or diffusion models, both of which suffer from high inference costs. In this paper, we propose a new approach capable of protein sequence and structure co-design, which iteratively translates both protein sequence and structure into the desired state from random initialization, based on context features given a priori. Our model consists of a trigonometry-aware encoder that reasons geometrical constraints and interactions from context features, and a roto-translation equivariant decoder that translates protein sequence and structure interdependently. Notably, all protein amino acids are updated in one shot in each translation step, which significantly accelerates the inference process. Experimental results across multiple tasks show that our model outperforms previous state-of-the-art baselines by a large margin, and is able to design proteins of high fidelity as regards both sequence and structure, with running time orders of magnitude less than sampling-based methods.

2023-02-01

ICLR.cc/2023/Conference (poster)

Design and Application of Adaptive Sparse Deep Echo State Network

Cuili Yang

Sheng Yang

Bing Li

The prediction of appliances energy consumption in building belongs to time series forecasting problem, which can be solved by echo state ne… (see more)twork (ESN). However, due to the randomly initialized inputs and reservoir, some redundant or irrelevant components are inevitably generated in original ESN. To solve this problem, the adaptive sparse deep echo state network (ASDESN) is proposed, in which the information is processed layer by layer. Firstly, the principal component analysis (PCA) layer is inserted to penalize the redundant projection transmitted between sub-reservoirs. Secondly, the coordinate descent based adaptive sparse learning method is proposed to generate the sparse output weights. Particularly, the designed adaptive threshold strategy is able to enlarge the sparsity of output weights as network depth increases. Moreover, the echo state property (ESP) of ASDESN is given to ensure its applications. The experiment results in both simulated benchmark and real appliances energy datasets illustrate that the proposed ASDESN outperforms other ESNs with higher prediction accuracy and stability.

2023-01-01

IEEE transactions on consumer electronics (published)

FusionRetro: Molecule Representation Fusion via In-Context Learning for Retrosynthetic Planning

Songtao Liu

Zhengkai Tu

Minkai Xu

Zuobai Zhang

Lu Lin

Rex Ying

Zhitao Ying

Peilin Zhao

Dinghao Wu

2023-01-01

ICML (published)

FusionRetro: Molecule Representation Fusion via Reaction Graph for Retrosynthetic Planning

Songtao Liu

Zhengkai Tu

Minkai Xu

Zuobai Zhang

Peilin Zhao

Rex Ying

Lu Lin

Dinghao Wu

Retrosynthetic planning is a fundamental problem in drug discovery and organic chemistry, which aims to ﬁnd a complete multi-step syntheti… (see more)c route from a set of starting materials to the target molecule, determining crucial process ﬂow in chemical production. Existing approaches combine single-step retrosynthesis models and search algorithms to ﬁnd synthetic routes. However, these approaches generally consider the two pieces in a decoupled manner, taking only the product as the input to predict the reactants per planning step and largely ignoring the important context information from other intermediates along the synthetic route. In this work, we perform a series of experiments to identify the limitations of this decoupled view and propose a novel retrosynthesis framework that also exploits context information for retrosynthetic planning. We view synthetic routes as reaction graphs, and propose to incorporate the context by three principled steps: encode molecules into embeddings, aggregate information over routes, and readout to predict reactants. The whole framework can be efﬁciently optimized in an end-to-end fashion. Comprehensive experiments show that by fusing in context information over routes, our model sig-niﬁcantly improves the performance of retrosyn-thetic planning over baselines that are not context-aware, especially for long synthetic routes.

A Group Symmetric Stochastic Differential Equation Model for Molecule Multi-modal Pretraining

Shengchao Liu

weitao Du

Zhi-Ming Ma

Hongyu Guo

Molecule pretraining has quickly become the go-to schema to boost the performance of AI-based drug discovery. Naturally, molecules can be re… (see more)presented as 2D topological graphs or 3D geometric point clouds. Although most existing pertaining methods focus on merely the single modality, recent research has shown that maximizing the mutual information (MI) between such two modalities enhances the molecule representation ability. Meanwhile, existing molecule multi-modal pretraining approaches approximate MI based on the representation space encoded from the topology and geometry, thus resulting in the loss of critical structural information of molecules. To address this issue, we propose MoleculeSDE. MoleculeSDE leverages group symmetric (e.g., SE(3)-equivariant and reflection-antisymmetric) stochastic differential equation models to generate the 3D geometries from 2D topologies, and vice versa, directly in the input space. It not only obtains tighter MI bound but also enables prosperous downstream tasks than the previous work. By comparing with 17 pretraining baselines, we empirically verify that MoleculeSDE can learn an expressive representation with state-of-the-art performance on 26 out of 32 downstream tasks.

2023-01-01

ICML (published)

Physics-Inspired Protein Encoder Pre-Training via Siamese Sequence-Structure Diffusion Trajectory Prediction

Zuobai Zhang

Minghao Xu

Aurelie Lozano

Vijil Chenthamarakshan

Payel Das

Pre-training methods on proteins are recently gaining interest, leveraging either protein sequences or structures, while modeling their join… (see more)t energy landscape is largely unexplored. In this work, inspired by the success of denoising diffusion models, we propose the DiffPreT approach to pre-train a protein encoder by sequence-structure multimodal diffusion modeling. DiffPreT guides the encoder to recover the native protein sequences and structures from the perturbed ones along the multimodal diffusion trajectory, which acquires the joint distribution of sequences and structures. Considering the essential protein conformational variations, we enhance DiffPreT by a physics-inspired method called Siamese Diffusion Trajectory Prediction ( SiamDiff ) to capture the correlation between different conformers of a protein. SiamDiff attains this goal by maximizing the mutual information between representations of diffusion trajectories of structurally-correlated conformers. We study the effectiveness of DiffPreT and SiamDiff on both atom-and residue-level structure-based protein understanding tasks. Experimental results show that the performance of DiffPreT is consistently competitive on all tasks, and SiamDiff achieves new state-of-the-art performance, considering the mean ranks on all tasks. The source code will be released upon acceptance.

2023-01-01

arXiv.org (preprint)

GraphCG: Unsupervised Discovery of Steerable Factors in Graphs

Shengchao Liu

Chengpeng Wang

Weili Nie

Hanchen Wang

Jiarui Lu

Bolei Zhou

Deep generative models have been extensively explored recently, especially for the graph data such as molecular graphs and point clouds. Yet… (see more), much less investigation has been carried out on understanding the learned latent space of deep graph generative models. Such understandings can open up a unified perspective and provide guidelines for essential tasks like controllable generation. In this paper, we first examine the representation space of the recent deep generative model trained for graph data, observing that the learned representation space is not perfectly disentangled. Based on this observation, we then propose an unsupervised method called GraphCG, which is model-agnostic and task-agnostic for discovering steerable factors in graph data. Specifically, GraphCG learns the semantic-rich directions via maximizing the corresponding mutual information, where the edited graph along the same direction will possess certain steerable factors. We conduct experiments on two types of graph data, molecular graphs and point clouds. Both the quantitative and qualitative results show the effectiveness of GraphCG for discovering steerable factors. The code will be public in the near future.

2022-11-22

NeurIPS.cc/2022/Workshop/GLFrontiers (published)

Flaky Performances when Pretraining on Relational Databases

Shengchao Liu

David Vazquez

Pierre-Andre Noel

2022-11-09

ArXiv (preprint)

arxiv.org

Implications of Topological Imbalance for Representation Learning on Biomedical Knowledge Graphs

Stephen Bonner

Ufuk Kirik

Ola Engkvist

Ian P Barrett

Adoption of recently developed methods from machine learning has given rise to creation of drug-discovery knowledge graphs (KGs) that utiliz… (see more)e the interconnected nature of the domain. Graph-based modelling of the data, combined with KG embedding (KGE) methods, are promising as they provide a more intuitive representation and are suitable for inference tasks such as predicting missing links. One common application is to produce ranked lists of genes for a given disease, where the rank is based on the perceived likelihood of association between the gene and the disease. It is thus critical that these predictions are not only pertinent but also biologically meaningful. However, KGs can be biased either directly due to the underlying data sources that are integrated or due to modelling choices in the construction of the graph, one consequence of which is that certain entities can get topologically overrepresented. We demonstrate the effect of these inherent structural imbalances, resulting in densely connected entities being highly ranked no matter the context. We provide support for this observation across different datasets, models as well as predictive tasks. Further, we present various graph perturbation experiments which yield more support to the observation that KGE models can be more influenced by the frequency of entities rather than any biological information encoded within the relations. Our results highlight the importance of data modelling choices, and emphasizes the need for practitioners to be mindful of these issues when interpreting model outputs and during KG composition.

2022-07-26

Briefings in Bioinformatics (published)

arxiv.org

Subgraph Retrieval Enhanced Model for Multi-hop Knowledge Base Question Answering

Jie Zhang

Xiaokang Zhang

Jifan Yu

Jie Tang

Cuiping Li

Hong Chen

2022-05-01

Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (published)