arXiv subjects

Jianzhu Ma

Publications and source records attributed to Jianzhu Ma.

At least 55 records · Page 3Linked to original sources

arXiv2023

3D Equivariant Diffusion for Target-Aware Molecule Generation and Affinity Prediction

Rich data and powerful machine learning models allow us to design drugs for a specific protein target \textit{in silico}. Recently, the inclusion of 3D structures during targeted drug design shows superior performance to other target-free models as the atomic interaction in the 3D space is explicitly modeled. However, current 3D target-aware models either rely on the voxelized atom densities or the autoregressive sampling process, which are not equivariant to rotation or easily violate geometric constraints resulting in unrealistic structures. In this work, we develop a 3D equivariant diffusion model to solve the above challenges. To achieve target-aware molecule design, our method learns a joint generative process of both continuous atom coordinates and categorical atom types with a SE(3)-equivariant network. Moreover, we show that our model can serve as an unsupervised feature extractor to estimate the binding affinity under proper parameterization, which provides an effective way for drug screening. To evaluate our model, we propose a comprehensive framework to evaluate the quality of sampled molecules from different dimensions. Empirical studies show our model could generate molecules with more realistic 3D structures and better affinities towards the protein targets, and improve binding affinity ranking and prediction without retraining.

q-bio.BM↗

arXiv2023

Scientific Computing with Diffractive Optical Neural Networks

Diffractive optical neural networks (DONNs) have been emerging as a high-throughput and energy-efficient hardware platform to perform all-optical machine learning (ML) in machine vision systems. However, the current demonstrated applications of DONNs are largely straightforward image classification tasks, which undermines the prospect of developing and utilizing such hardware for other ML applications. Here, we numerically and experimentally demonstrate the deployment of an all-optical reconfigurable DONNs system for scientific computing, including guiding two-dimensional quantum material synthesis, predicting the properties of nanomaterials and small molecular cancer drugs, predicting the device response of nanopatterned integrated photonic power splitters, and the dynamic stabilization of an inverted pendulum with reinforcement learning. Despite a large variety of input data structures, we develop a universal feature engineering approach to convert categorical input features to the images that can be processed in the DONNs system. Our results open up new opportunities of employing DONNs systems for a broad range of ML applications.

cs.LG↗

arXiv2022

Efficient Meta Reinforcement Learning for Preference-based Fast Adaptation

Learning new task-specific skills from a few trials is a fundamental challenge for artificial intelligence. Meta reinforcement learning (meta-RL) tackles this problem by learning transferable policies that support few-shot adaptation to unseen tasks. Despite recent advances in meta-RL, most existing methods require the access to the environmental reward function of new tasks to infer the task objective, which is not realistic in many practical applications. To bridge this gap, we study the problem of few-shot adaptation in the context of human-in-the-loop reinforcement learning. We develop a meta-RL algorithm that enables fast policy adaptation with preference-based feedback. The agent can adapt to new tasks by querying human's preference between behavior trajectories instead of using per-step numeric rewards. By extending techniques from information theory, our approach can design query sequences to maximize the information gain from human interactions while tolerating the inherent error of non-expert human oracle. In experiments, we extensively evaluate our method, Adaptation with Noisy OracLE (ANOLE), on a variety of meta-RL benchmark tasks and demonstrate substantial improvement over baseline algorithms in terms of both feedback efficiency and error tolerance.

cs.LG↗

arXiv2022

A 3D Generative Model for Structure-Based Drug Design

We study a fundamental problem in structure-based drug design -- generating molecules that bind to specific protein binding sites. While we have witnessed the great success of deep generative models in drug design, the existing methods are mostly string-based or graph-based. They are limited by the lack of spatial information and thus unable to be applied to structure-based design tasks. Particularly, such models have no or little knowledge of how molecules interact with their target proteins exactly in 3D space. In this paper, we propose a 3D generative model that generates molecules given a designated 3D protein binding site. Specifically, given a binding site as the 3D context, our model estimates the probability density of atom's occurrences in 3D space -- positions that are more likely to have atoms will be assigned higher probability. To generate 3D molecules, we propose an auto-regressive sampling scheme -- atoms are sampled sequentially from the learned distribution until there is no room for new atoms. Combined with this sampling scheme, our model can generate valid and diverse molecules, which could be applicable to various structure-based molecular design tasks such as molecule sampling and linker design. Experimental results demonstrate that molecules sampled from our model exhibit high binding affinity to specific targets and good drug properties such as drug-likeness even if the model is not explicitly optimized for them.

q-bio.BM↗

arXiv2022

3DLinker: An E(3) Equivariant Variational Autoencoder for Molecular Linker Design

Deep learning has achieved tremendous success in designing novel chemical compounds with desirable pharmaceutical properties. In this work, we focus on a new type of drug design problem -- generating a small "linker" to physically attach two independent molecules with their distinct functions. The main computational challenges include: 1) the generation of linkers is conditional on the two given molecules, in contrast to generating full molecules from scratch in previous works; 2) linkers heavily depend on the anchor atoms of the two molecules to be connected, which are not known beforehand; 3) 3D structures and orientations of the molecules need to be considered to avoid atom clashes, for which equivariance to E(3) group are necessary. To address these problems, we propose a conditional generative model, named 3DLinker, which is able to predict anchor atoms and jointly generate linker graphs and their 3D structures based on an E(3) equivariant graph variational autoencoder. So far as we know, there are no previous models that could achieve this task. We compare our model with multiple conditional generative models modified from other molecular design tasks and find that our model has a significantly higher rate in recovering molecular graphs, and more importantly, accurately predicting the 3D coordinates of all the atoms.

cs.LG↗

arXiv2022

Equivariant Point Cloud Analysis via Learning Orientations for Message Passing

Equivariance has been a long-standing concern in various fields ranging from computer vision to physical modeling. Most previous methods struggle with generality, simplicity, and expressiveness -- some are designed ad hoc for specific data types, some are too complex to be accessible, and some sacrifice flexible transformations. In this work, we propose a novel and simple framework to achieve equivariance for point cloud analysis based on the message passing (graph neural network) scheme. We find the equivariant property could be obtained by introducing an orientation for each point to decouple the relative position for each point from the global pose of the entire point cloud. Therefore, we extend current message passing networks with a module that learns orientations for each point. Before aggregating information from the neighbors of a point, the networks transforms the neighbors' coordinates based on the point's learned orientations. We provide formal proofs to show the equivariance of the proposed framework. Empirically, we demonstrate that our proposed method is competitive on both point cloud analysis and physical modeling tasks. Code is available at https://github.com/luost26/Equivariant-OrientedMP .

cs.CV↗

arXiv2022

Provable Constrained Stochastic Convex Optimization with XOR-Projected Gradient Descent

Provably solving stochastic convex optimization problems with constraints is essential for various problems in science, business, and statistics. Recently proposed XOR-Stochastic Gradient Descent (XOR-SGD) provides a convergence rate guarantee solving the constraints-free version of the problem by leveraging XOR-Sampling. However, the task becomes more difficult when additional equality and inequality constraints are needed to be satisfied. Here we propose XOR-PGD, a novel algorithm based on Projected Gradient Descent (PGD) coupled with the XOR sampler, which is guaranteed to solve the constrained stochastic convex optimization problem still in linear convergence rate by choosing proper step size. We show on both synthetic stochastic inventory management and real-world road network design problems that the rate of constraints satisfaction of the solutions optimized by XOR-PGD is $10\%$ more than the competing approaches in a very large searching space. The improved XOR-PGD algorithm is demonstrated to be more accurate and efficient than both XOR-SGD and SGD coupled with MCMC based samplers. It is also shown to be more scalable with respect to the number of samples and processor cores via experiments with large dimensions.

math.OC↗

arXiv2022

Device-system Co-design of Photonic Neuromorphic Processor using Reinforcement Learning

The incorporation of high-performance optoelectronic devices into photonic neuromorphic processors can substantially accelerate computationally intensive operations in machine learning (ML) algorithms. However, the conventional device design wisdom is disconnected with system optimization. We report a device-system co-design methodology to optimize a free-space optical general matrix multiplication (GEMM) hardware accelerator by engineering a spatially reconfigurable array made from chalcogenide phase change materials. With a highly-parallelized hardware emulator constructed based on experimental information, we demonstrate the design of unit device by optimizing GEMM calculation accuracy via reinforcement learning, including deep Q-learning neural network, Bayesian optimization, and their cascaded approach, which show a clear correlation between system performance metrics and physical device specifications. Furthermore, we employ physics-aware training approaches to deploy optimized hardware to the tasks of image classification, materials discovery, and a closed-loop design of optical ML accelerators. The demonstrated framework offers insights into the co-design of optoelectronic devices and systems with reduced human-supervision and domain-knowledge barriers.

cs.ET↗

arXiv2022

Neural Predicting Higher-order Patterns in Temporal Networks

Dynamic systems that consist of a set of interacting elements can be abstracted as temporal networks. Recently, higher-order patterns that involve multiple interacting nodes have been found crucial to indicate domain-specific laws of different temporal networks. This posts us the challenge of designing more sophisticated hypergraph models for these higher-order patterns and the associated new learning algorithms. Here, we propose the first model, named HIT, for full-spectrum higher-order pattern prediction in temporal hypergraphs. Particularly, we focus on predicting three types of common but important interaction patterns involving three interacting elements in temporal networks, which could be extended to even higher-order patterns. HIT extracts the structural representation of a node triplet of interest on the temporal hypergraph and uses it to tell what type of, when, and why the interaction expansion could happen in this triplet. HIT could achieve significant improvement (averaged 20% AUC gain to identify the interaction type, uniformly more accurate time estimation) compared to both heuristic and other neural-network-based baselines on 5 real-world large temporal hypergraphs. Moreover, HIT provides a certain degree of interpretability by identifying the most discriminatory structural features on the temporal hypergraphs for predicting different higher-order patterns.

cs.SI↗

arXiv2021

Contrastive Cycle Adversarial Autoencoders for Single-cell Multi-omics Alignment and Integration

Muilti-modality data are ubiquitous in biology, especially that we have entered the multi-omics era, when we can measure the same biological object (cell) from different aspects (omics) to provide a more comprehensive insight into the cellular system. When dealing with such multi-omics data, the first step is to determine the correspondence among different modalities. In other words, we should match data from different spaces corresponding to the same object. This problem is particularly challenging in the single-cell multi-omics scenario because such data are very sparse with extremely high dimensions. Secondly, matched single-cell multi-omics data are rare and hard to collect. Furthermore, due to the limitations of the experimental environment, the data are usually highly noisy. To promote the single-cell multi-omics research, we overcome the above challenges, proposing a novel framework to align and integrate single-cell RNA-seq data and single-cell ATAC-seq data. Our approach can efficiently map the above data with high sparsity and noise from different spaces to a low-dimensional manifold in a unified space, making the downstream alignment and integration straightforward. Compared with the other state-of-the-art methods, our method performs better in both simulated and real single-cell data. The proposed method is helpful for the single-cell multi-omics research. The improvement for integration on the simulated data is significant.

q-bio.GN↗

arXiv2021

Quantifying Spatial Homogeneity of Urban Road Networks via Graph Neural Networks

Quantifying the topological similarities of different parts of urban road networks (URNs) enables us to understand the urban growth patterns. While conventional statistics provide useful information about characteristics of either a single node's direct neighbors or the entire network, such metrics fail to measure the similarities of subnetworks considering local indirect neighborhood relationships. In this study, we propose a graph-based machine-learning method to quantify the spatial homogeneity of subnetworks. We apply the method to 11,790 urban road networks across 30 cities worldwide to measure the spatial homogeneity of road networks within each city and across different cities. We find that intra-city spatial homogeneity is highly associated with socioeconomic statuses such as GDP and population growth. Moreover, inter-city spatial homogeneity obtained by transferring the model across different cities, reveals the inter-city similarity of urban network structures originating in Europe, passed on to cities in the US and Asia. Socioeconomic development and inter-city similarity revealed using our method can be leveraged to understand and transfer insights across cities. It also enables us to address urban policy challenges including network planning in rapidly urbanizing areas and combating regional inequality.

physics.soc-ph↗

arXiv2021

Fast Projection onto the Capped Simplex with Applications to Sparse Regression in Bioinformatics

We consider the problem of projecting a vector onto the so-called k-capped simplex, which is a hyper-cube cut by a hyperplane. For an n-dimensional input vector with bounded elements, we found that a simple algorithm based on Newton's method is able to solve the projection problem to high precision with a complexity roughly about O(n), which has a much lower computational cost compared with the existing sorting-based methods proposed in the literature. We provide a theory for partial explanation and justification of the method. We demonstrate that the proposed algorithm can produce a solution of the projection problem with high precision on large scale datasets, and the algorithm is able to significantly outperform the state-of-the-art methods in terms of runtime (about 6-8 times faster than a commercial software with respect to CPU time for input vector with 1 million variables or more). We further illustrate the effectiveness of the proposed algorithm on solving sparse regression in a bioinformatics problem. Empirical results on the GWAS dataset (with 1,500,000 single-nucleotide polymorphisms) show that, when using the proposed method to accelerate the Projected Quasi-Newton (PQN) method, the accelerated PQN algorithm is able to handle huge-scale regression problem and it is more efficient (about 3-6 times faster) than the current state-of-the-art methods.

math.OC↗

arXiv2021

Physics-Guided and Physics-Explainable Recurrent Neural Network for Time Dynamics in Optical Resonances

Understanding the time evolution of physical systems is crucial to revealing fundamental characteristics that are hidden in frequency domain. In optical science, high-quality resonance cavities and enhanced interactions with matters are at the heart of modern quantum technologies. However, capturing their time dynamics in real-world scenarios suffers from long data acquisition and low analysis accuracy due to slow convergence and limited time window. Here, we report a physics-guided and physics-explainable recurrent neural network to precisely forecast the time-domain response of resonance features with the shortest acquired input sequence being 7\% of full length, and to infer corresponding resonance frequencies. The model is trained in a two-step multi-fidelity framework for high-accuracy forecast, where the first step is based on a large amount of low-fidelity physical-model-generated synthetic data and second step involves a small set of high-fidelity application-oriented observational data. Through both simulations and experiments, we demonstrate that the model is universally applicable to a wide range of resonances, including dielectric metasurfaces, graphene plasmonics, and ultrastrongly coupled Landau polaritons, where our model accurately captures small signal features and learns essential physical quantities. The demonstrated machine learning algorithm offers a new way to accelerate the exploration of physical phenomena and the design of devices under resonance-enhanced light-matter interaction.

physics.optics↗

arXiv2021

Breaking the Limit of Graph Neural Networks by Improving the Assortativity of Graphs with Local Mixing Patterns

Graph neural networks (GNNs) have achieved tremendous success on multiple graph-based learning tasks by fusing network structure and node features. Modern GNN models are built upon iterative aggregation of neighbor's/proximity features by message passing. Its prediction performance has been shown to be strongly bounded by assortative mixing in the graph, a key property wherein nodes with similar attributes mix/connect with each other. We observe that real world networks exhibit heterogeneous or diverse mixing patterns and the conventional global measurement of assortativity, such as global assortativity coefficient, may not be a representative statistic in quantifying this mixing. We adopt a generalized concept, node-level assortativity, one that is based at the node level to better represent the diverse patterns and accurately quantify the learnability of GNNs. We find that the prediction performance of a wide range of GNN models is highly correlated with the node level assortativity. To break this limit, in this work, we focus on transforming the input graph into a computation graph which contains both proximity and structural information as distinct type of edges. The resulted multi-relational graph has an enhanced level of assortativity and, more importantly, preserves rich information from the original graph. We then propose to run GNNs on this computation graph and show that adaptively choosing between structure and proximity leads to improved performance under diverse mixing. Empirically, we show the benefits of adopting our transformation framework for semi-supervised node classification task on a variety of real world graph learning benchmarks.

cs.LG↗

arXiv2015

Protein secondary structure prediction using deep convolutional neural fields

Protein secondary structure (SS) prediction is important for studying protein structure and function. When only the sequence (profile) information is used as input feature, currently the best predictors can obtain ~80% Q3 accuracy, which has not been improved in the past decade. Here we present DeepCNF (Deep Convolutional Neural Fields) for protein SS prediction. DeepCNF is a Deep Learning extension of Conditional Neural Fields (CNF), which is an integration of Conditional Random Fields (CRF) and shallow neural networks. DeepCNF can model not only complex sequence-structure relationship by a deep hierarchical architecture, but also interdependency between adjacent SS labels, so it is much more powerful than CNF. Experimental results show that DeepCNF can obtain ~84% Q3 accuracy, ~85% SOV score, and ~72% Q8 accuracy, respectively, on the CASP and CAMEO test proteins, greatly outperforming currently popular predictors. As a general framework, DeepCNF can be used to predict other protein structure properties such as contact number, disorder regions, and solvent accessibility.

q-bio.BM↗

arXiv2015

Predicting diverse M-best protein contact maps

Protein contacts contain important information for protein structure and functional study, but contact prediction from sequence information remains very challenging. Recently evolutionary coupling (EC) analysis, which predicts contacts by detecting co-evolved residues (or columns) in a multiple sequence alignment (MSA), has made good progress due to better statistical assessment techniques and high-throughput sequencing. Existing EC analysis methods predict only a single contact map for a given protein, which may have low accuracy especially when the protein under prediction does not have a large number of sequence homologs. Analogous to ab initio folding that usually predicts a few possible 3D models for a given protein sequence, this paper presents a novel structure learning method that can predict a set of diverse contact maps for a given protein sequence, in which the best solution usually has much better accuracy than the first one. Our experimental tests show that for many test proteins, the best out of 5 solutions generated by our method has accuracy at least 0.1 better than the first one when the top L/5 or L/10 (L is the sequence length) predicted long-range contacts are evaluated, especially for protein families with a small number of sequence homologs. Our best solutions also have better quality than those generated by the two popular EC methods Evfold and PSICOV.

q-bio.QM↗

arXiv2015

Protein Structure Prediction by Protein Alignments

Proteins are the basic building blocks of life. They usually perform functions by folding to a particular structure. Understanding the folding process could help the researchers to understand the functions of proteins and could also help to develop supplemental proteins for people with deficiencies and gain more insight into diseases associated with troublesome folding proteins. Experimental methods are both expensive and time consuming. In this thesis I introduce a new machine learning based method to predict the protein structure. The new method improves the performance from two directions: creating accurate protein alignments and predicting accurate protein contacts. First, I present an alignment framework MRFalign which goes beyond state-of-the-art methods and uses Markov Random Fields to model a protein family and align two proteins by aligning two MRFs together. Compared to other methods, that can only model local-range residue correlation, MRFs can model long-range residue interactions and thus, encodes global information in a protein. Secondly, I present a Group Graphical Lasso method for contact prediction that integrates joint multi-family Evolutionary Coupling analysis and supervised learning to improve accuracy on proteins without many sequence homologs. Different from single-family EC analysis that uses residue co-evolution information in only the target protein family, our joint EC analysis uses residue co-evolution in both the target family and its related families, which may have divergent sequences but similar folds. Our method can also integrate supervised learning methods to further improve accuracy. We evaluate the performance of both methods including each of its components on large public benchmarks. Experiments show that our methods can achieve better accuracy than existing state-of-the-art methods under all the measurements on most of the protein classes.

cs.CE↗

arXiv2015

Bermuda: Bidirectional de novo assembly of transcripts with new insights for handling uneven coverage

Motivation: RNA-seq has made feasible the analysis of a whole set of expressed mRNAs. Mapping-based assembly of RNA-seq reads sometimes is infeasible due to lack of high-quality references. However, de novo assembly is very challenging due to uneven expression levels among transcripts and also the read coverage variation within a single transcript. Existing methods either apply de Bruijn graphs of single-sized k-mers to assemble the full set of transcripts, or conduct multiple runs of assembly, but still apply graphs of single-sized k-mers at each run. However, a single k-mer size is not suitable for all the regions of the transcripts with varied coverage. Contribution: This paper presents a de novo assembler Bermuda with new insights for handling uneven coverage. Opposed to existing methods that use a single k-mer size for all the transcripts in each run of assembly, Bermuda self-adaptively uses a few k-mer sizes to assemble different regions of a single transcript according to their local coverage. As such, Bermuda can deal with uneven expression levels and coverage not only among transcripts, but also within a single transcript. Extensive tests show that Bermuda outperforms popular de novo assemblers in reconstructing unevenly-expressed transcripts with longer length, better contiguity and lower redundancy. Further, Bermuda is computationally efficient with moderate memory consumption.

q-bio.GN↗