Multi-View Multi-Instance Multi-Label Learning Based on Collaborative Matrix Factorization

Bioinformatics Research and Applications

Zhang

et al. 2020

Self Cite

Alternative splicing enables a gene spliced into different isoforms, which are closely related with diverse developmental abnormalities. Identifying the isoform-disease associations helps to uncover the underlying pathology of various complex diseases, and to develop precise treatments and drugs for these diseases. Although many approaches have been proposed for predicting gene-disease associations and isoform functions, few efforts have been made toward predicting isoform-disease associations in large-scale, the main bottleneck is the lack of ground-truth isoform-disease associations. To bridge this gap, we propose a multiinstance learning inspired computational approach called IDAPred to fuse genomics and transcriptomics data for isoform-disease association prediction. Given the bag-instance relationship between gene and its spliced isoforms, IDAPred introduces a dispatch and aggregation term to dispatch gene-disease associations to individual isoforms, and reversely aggregate these dispatched associations to affiliated genes. Next, it fuses different genomics and transcriptomics data to replenish gene-disease associations and to induce a linear classifier for predicting isoform-disease associations in a coherent way. In addition, to alleviate the bias toward observed gene-disease associations, it adds a regularization term to differentiate the currently observed associations from the unobserved (potential) ones. Experimental results show that IDAPred significantly outperforms the related state-of-the-art methods.

Section: Methodsmentioning

confidence: 99%

Isoform-Disease Association Prediction by Data Fusion

Huang

Bioinformatics Research and Applications

Zhang

et al. 2020

Self Cite

“…Both the inter(intra)-relation between bags and those between instances of a bag can be used to improve the performance of M3L [37], [3]. To mine more shared information, the third term in Eq.…”

Section: B Extracting the Shared And Individual Informationmentioning

confidence: 99%

“…As a framework for modeling complex objects, Multiview Multi-instance Multi-label Learning (M3L) has attracted increasing interest in various applications [1], [2], [3], such as video annotation [4] and functional genomics [5], [6], [7]. In M3L, each object (or bag) contains one or more instances, is represented with different feature views, and is annotated with a set of non-exclusive semantic labels.…”

Section: Introductionmentioning

confidence: 99%

Multiview Multi-Instance Multilabel Active Learning

Xing

IEEE Trans. Neural Netw. Learning Syst.

et al. 2022

Self Cite

Multi-view multi-instance multi-label learning (M3L) is a framework for modeling complex objects. In this framework, each object (or bag) contains one or more instances, is represented with different feature views, and simultaneously annotated with a set of non-exclusive semantic labels. Given the multiplicity of the studied objects, traditional M3L methods generally demand a large number of labeled bags to train a predictive model to annotate bags (or instances) with semantic labels. However, annotating sufficient bags is very expensive and often impractical. In this paper, we present an Active learning based M3L approach (M3AL) to reduce the labeling costs of bags and to improve the performance as much as possible. M3AL firstly adapts the multi-view self-representation learning to evacuate the shared and individual information of bags and to learn the shared/individual similarities between bags across/within views. Next, to avoid scrutinizing all the possible labels, M3AL introduces a new query strategy that leverages the shared and individual information, and the diverse instance distribution of bags across views, to select the most informative bag-label pair for query. Experimental studies on benchmark datasets show that M3AL can significantly reduce the query costs while achieve a better performance than other related competitive methods at the same cost.

“…Multi‐instance learning (MIL) has been extensively studied and applied in diverse domains, such as image annotations, 1‐5 functional genomics, 6‐10 text mining 11,12 and so on. MIL models complex bags (samples) that are made of diverse instances (subsamples), in which the labels of bags are available but those of individual instances are unspecified.…”

Section: Introductionmentioning

confidence: 99%

Imbalance deep multi‐instance learning for predicting isoform–isoform interactions

Zeng

et al. 2021

Int J Intell Syst

Self Cite

Multi‐instance learning (MIL) can model complex bags (samples) that are further made of diverse instances (subsamples). In typical MIL, the labels of bags are known while those of individual instances are unknown and to be specified. In this paper we propose an imbalanced deep multi‐instance learning approach (IDMIL‐III) and apply it to predict genome‐wide isoform–isoform interactions (IIIs). This prediction task is crucial for precisely understanding the interactome between proteoforms and to reveal their functional diversity. The current solutions typically formulate the prediction of IIIs as a MIL problem by pairing two genes as a “bag” and any two isoforms spliced from these two genes as “instances.” The key instances (interacting isoform pairs) trigger the label of the positive (interacting) gene bags, which is important for identifying the IIIs. Furthermore, the prediction task was simplified as a balanced classification problem, which in practice is a rather imbalanced one. To address these issues, IDMIL‐III fuses RNA‐seq, nucleotide sequence, amino acid sequence and exon array data, and further introduces a novel loss function to separately model the loss of positive pairs and of negative pairs, and thus to avoid the expected loss dominated by majority negative pairs. In addition, it includes an attention strategy to identify positive isoform pairs from a positive gene bag. Extensive experimental results prove the effectiveness of IDMIL‐III on predicting IIIs. Particularly, IDMIL‐III achieves an F1 value as 95.4%, at least 3.8% higher than those of competitive methods at the gene‐level; and obtains an F1 as 29.8%, at least 2.4% higher than the state‐of‐the‐art methods at the isoform‐level. The code of IDMIL‐III is available at http://mlda.swu.edu.cn/codes.php?name=IDMIL-III.