Proceedings of the ... SIAM International Conference on Data Mining. SIAM International Conference on Data Mining最新文献

Large pre-trained models have been extensively explored for numerous biomedical tasks. However, the diversity and complexity of biological systems often make zero-shot learning in a new context challenging. In many instances, the budget allows for the acquisition of a small number of labeled data through experiments for few-shot learning. Yet, the methodology for selecting the optimal set of samples for these experiments remains underexplored. In this work, we present an application focused on drug-induced gene expression prediction to demonstrate a data-driven approach for facilitating sample selection. We developed a system named AnchorDrug, which predicts drug-induced gene expression changes in new cell lines after fine-tuning with experimental data from a limited number of drugs. Initially, we built a pre-trained model with a large dataset of drug-induced gene expressions. We then adopted active learning to identify an optimal set of drugs (i.e. anchor drugs) for experiments, aiming to ensure that the experimental data used for subsequent fine-tuning would maximize model performance. Several acquisition functions are customized and incorporated into our pipeline. Compared with knowledge-based drug selection, our customized active learning methods proved more effective in selecting anchor drugs. A model trained using data from anchor drugs can even perform better than that trained using all available data in certain scenarios. We further provided insights into the reasons behind its superior performance. Our system is designed to mimic real-world scenarios, enabling its easy application to real biomedical research projects.

Environmental ecosystems exhibit complex and evolving dynamics over time, making the modeling of non-stationary processes critically important. However, traditional methods often rely on static models trained on entire datasets, failing to capture the non-stationary and drastically fluctuating characteristics. Dynamically adjusting models to evolving data is challenging, as they can easily either lag behind new trends or overfit newly received data. To address these challenges, we propose Domain-Adaptive Continual Meta-Learning (DACM) method, aiming to automatically detect distribution shifts and adapt to newly emergent domains. In particular, while DACM continuously explores the sequential temporal data, it also exploits historical data that are similar in distribution to the current observations. By striking a balance between temporal exploration and distributional exploitation, DACM quickly adjusts the model to stay up-to-date with new trends while maintaining generalization ability to data with similar distributions. We demonstrate the effectiveness of DACM on a real-world water temperature prediction dataset, where it outperforms diverse baseline models and shows strong adaptability and predictive performance in non-stationary environments.

Health risk prediction aims to forecast the potential health risks that patients may face using their historical Electronic Health Records (EHR). Although several effective models have developed, data insufficiency is a key issue undermining their effectiveness. Various data generation and augmentation methods have been introduced to mitigate this issue by expanding the size of the training data set through learning underlying data distributions. However, the performance of these methods is often limited due to their task-unrelated design. To address these shortcomings, this paper introduces a novel, end-to-end diffusion-based risk prediction model, named MedDiffusion. It enhances risk prediction performance by creating synthetic patient data during training to enlarge sample space. Furthermore, MedDiffusion discerns hidden relationships between patient visits using a step-wise attention mechanism, enabling the model to automatically retain the most vital information for generating high-quality data. Experimental evaluation on four real-world medical datasets demonstrates that MedDiffusion outperforms 14 cutting-edge baselines in terms of PR-AUC, F1, and Cohen's Kappa. We also conduct ablation studies and benchmark our model against GAN-based alternatives to further validate the rationality and adaptability of our model design. Additionally, we analyze generated data to offer fresh insights into the model's interpretability. The source code is available via https://shorturl.at/aerT0.

The widespread adoption of Electronic Health Record (EHR) systems in healthcare institutes has generated vast amounts of medical data, offering significant opportunities for improving healthcare services through deep learning techniques. However, the complex and diverse modalities and feature structures in real-world EHR data pose great challenges for deep learning model design. To address the multi-modality challenge in EHR data, current approaches primarily rely on hand-crafted model architectures based on intuition and empirical experiences, leading to sub-optimal model architectures and limited performance. Therefore, to automate the process of model design for mining EHR data, we propose a novel neural architecture search (NAS) framework named AutoFM, which can automatically search for the optimal model architectures for encoding diverse input modalities and fusion strategies. We conduct thorough experiments on real-world multi-modal EHR data and prediction tasks, and the results demonstrate that our framework not only achieves significant performance improvement over existing state-of-the-art methods but also discovers meaningful network architectures effectively.

De novo molecular design is a key challenge in drug discovery due to the complexity of chemical space. With the availability of molecular datasets and advances in machine learning, many deep generative models are proposed for generating novel molecules with desired properties. However, most of the existing models focus only on molecular distribution learning and target-based molecular design, thereby hindering their potentials in real-world applications. In drug discovery, phenotypic molecular design has advantages over target-based molecular design, especially in first-in-class drug discovery. In this work, we propose the first deep graph generative model (FAME) targeting phenotypic molecular design, in particular gene expression-based molecular design. FAME leverages a conditional variational autoencoder framework to learn the conditional distribution generating molecules from gene expression profiles. However, this distribution is difficult to learn due to the complexity of the molecular space and the noisy phenomenon in gene expression data. To tackle these issues, a gene expression denoising (GED) model that employs contrastive objective function is first proposed to reduce noise from gene expression data. FAME is then designed to treat molecules as the sequences of fragments and learn to generate these fragments in autoregressive manner. By leveraging this fragment-based generation strategy and the denoised gene expression profiles, FAME can generate novel molecules with a high validity rate and desired biological activity. The experimental results show that FAME outperforms existing methods including both SMILES-based and graph-based deep generative models for phenotypic molecular design. Furthermore, the effective mechanism for reducing noise in gene expression data proposed in our study can be applied to omics data modeling in general for facilitating phenotypic drug discovery.

We propose a novel framework for combining datasets via alignment of their intrinsic geometry. This alignment can be used to fuse data originating from disparate modalities, or to correct batch effects while preserving intrinsic data structure. Importantly, we do not assume any pointwise correspondence between datasets, but instead rely on correspondence between a (possibly unknown) subset of data features. We leverage this assumption to construct an isometric alignment between the data. This alignment is obtained by relating the expansion of data features in harmonics derived from diffusion operators defined over each dataset. These expansions encode each feature as a function of the data geometry. We use this to relate the diffusion coordinates of each dataset through our assumption of partial feature correspondence. Then, a unified diffusion geometry is constructed over the aligned data, which can also be used to correct the original data measurements. We demonstrate our method on several datasets, showing in particular its effectiveness in biological applications including fusion of single-cell RNA sequencing (scRNA-seq) and single-cell ATAC sequencing (scATAC-seq) data measured on the same population of cells, and removal of batch effect between biological samples.

Integrative analysis jointly analyzes multiple data sets to overcome curse of dimensionality. It can detect important but weak signals by jointly selecting features for all data sets, but unfortunately the sets of important features are not always the same for all data sets. Variations which allows heterogeneous sparsity structure-a subset of data sets can have a zero coefficient for a selected feature-have been proposed, but it compromises the effect of integrative analysis recalling the problem of losing weak important signals. We propose a new integrative analysis approach which not only aggregates weak important signals well in homogeneity setting but also substantially alleviates the problem of losing weak important signals in heterogeneity setting. Our approach exploits a priori known graphical structure of features by forcing joint selection of adjacent features, and integrating such information over multiple data sets can increase the power while taking into account the heterogeneity across data sets. We confirm the problem of existing approaches and demonstrate the superiority of our method through a simulation study and an application to gene expression data from ADNI.

Learning of classification models in practice often relies on human annotation effort in which humans assign class labels to data instances. As this process can be very time-consuming and costly, finding effective ways to reduce the annotation cost becomes critical for building such models. To solve this problem, instead of soliciting instance-based annotation we explore region-based annotation as the human feedback. A region is defined as a hyper-cubic subspace of the input space X and it covers a subpopulation of data instances that fall into this region. Each region is labeled with a number in [0,1] (in binary classification setting), representing a human estimate of the positive (or negative) class proportion in the subpopulation. To quickly discover pure regions (in terms of class proportion) in the data, we have developed a novel active learning framework that constructs regions in a hierarchical and adaptive way. Hierarchical means that regions are incrementally built into a hierarchical tree, which is done by repeatedly splitting the input space. Adaptive means that our framework can adaptively choose the best heuristic for each of the region splits. Through experiments on numerous datasets we demonstrate that our framework can identify pure regions in very few region queries. Thus our approach is shown to be effective in learning classification models from very limited human feedback.

Tensor factorization is a methodology that is applied in a variety of fields, ranging from climate modeling to medical informatics. A tensor is an n-way array that captures the relationship between n objects. These multiway arrays can be factored to study the underlying bases present in the data. Two challenges arising in tensor factorization are 1) the resulting factors can be noisy and highly overlapping with one another and 2) they may not map to insights within a domain. However, incorporating supervision to increase the number of insightful factors can be costly in terms of the time and domain expertise necessary for gathering labels or domain-specific constraints. To meet these challenges, we introduce CANDECOMP/PARAFAC (CP) tensor factorization with Cannot-Link Intermode Constraints (CP-CLIC), a framework that achieves succinct, diverse, interpretable factors. This is accomplished by gradually learning constraints that are verified with auxiliary information during the decomposition process. We demonstrate CP-CLIC's potential to extract sparse, diverse, and interpretable factors through experiments on simulated data and a real-world application in medical informatics.

Heterogeneous information networks (HINs) are ubiquitous in real-world applications. Due to the heterogeneity in HINs, the typed edges may not fully align with each other. In order to capture the semantic subtlety, we propose the concept of aspects with each aspect being a unit representing one underlying semantic facet. Meanwhile, network embedding has emerged as a powerful method for learning network representation, where the learned embedding can be used as features in various downstream applications. Therefore, we are motivated to propose a novel embedding learning framework-ASPEM-to preserve the semantic information in HINs based on multiple aspects. Instead of preserving information of the network in one semantic space, ASPEM encapsulates information regarding each aspect individually. In order to select aspects for embedding purpose, we further devise a solution for ASPEM based on dataset-wide statistics. To corroborate the efficacy of ASPEM, we conducted experiments on two real-words datasets with two types of applications-classification and link prediction. Experiment results demonstrate that ASPEM can outperform baseline network embedding learning methods by considering multiple aspects, where the aspects can be selected from the given HIN in an unsupervised manner.