MRM-Lasso: A Sparse Multiview Feature Selection Method via Low-Rank Analysis

Yang, Wanqi; Gao, Yang; Shi, Yinghuan; Cao, Longbing

doi:10.1109/tnnls.2015.2396937

Cited by 60 publications

(19 citation statements)

References 35 publications

(48 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The least absolute shrinkage and selection operator (LASSO) method can reduce the dimensionality of high-dimensional data and obtain a better-fitting model. The LASSO Cox regression model was used to identify the ideal coefficients for each prognostic signature [37,38]. Multivariate Cox regression analysis was employed in the most important survival-related AS events that were selected from each AS type to establish a prognostic signature (PS).…”

Section: Prognostic Signatures For Alternative Splicing Events In Germentioning

confidence: 99%

Prognostic Value and Potential Regulatory Mechanism of Alternative Splicing in Geriatric Breast Cancer

Wang

et al. 2020

Genes

View full text Add to dashboard Cite

Breast cancer has the highest mortality and morbidity among women, especially in elderly women over 60 years old. Abnormal alternative splicing (AS) events are associated with the occurrence and development of geriatric breast cancer (GBC), yet strong evidence is lacking for the prognostic value of AS in GBC and the regulatory network of AS in GBC, which may highlight the mechanism through which AS contributes to GBC. In the present study, we obtained splicing event information (SpliceSeq) and clinical information for GBC from The Cancer Genome Atlas, and we constructed a GBC prognosis model based on AS events to predict the survival outcomes of GBC. Kaplan–Meier analysis was conducted to evaluate the predictive accuracy among different molecular subtypes of GBC. We conducted enrichment analysis and constructed a splicing network between AS and the splicing factor (SF) to examine the possible regulatory mechanisms of AS in GBC. We constructed eight prognostic signatures with very high statistical accuracy in predicting GBC survival outcomes from 45,421 AS events of 10,480 genes detected in 462 GBC patients; the prognostic model based on exon skip (ES) events had the highest accuracy, indicating its significant value in GBC prognosis. The constructed regulatory SF–AS network may explain the potential regulatory mechanism between SF and AS, which may be the mechanism through which AS events contribute to GBC survival outcomes. The findings confirm that AS events have a significant prognostic value in GBC, and we found a few effective prognostic signatures. We also hypothesized the mechanism underlying AS in GBC and discovered a potential regulatory mechanism between SF and AS.

show abstract

Section: Prognostic Signatures For Alternative Splicing Events In Germentioning

confidence: 99%

Prognostic Value and Potential Regulatory Mechanism of Alternative Splicing in Geriatric Breast Cancer

Wang

et al. 2020

Genes

View full text Add to dashboard Cite

show abstract

“…For convenience, we use ∈ × denote i-th column in connectivity matrix S, which characters the connections of region Oi with respect to other brain regions. Also, we arrange all Pearson's correlation values into a N × N matrix = { | = 1, ⋯ , } Instead of calculating the connectivity sij just based on Pearson's correlation c(xi,xj) between observed BOLD signals xi and xj, we optimize the connectivity matrix S by integrating the above three criteria: argmin S || − || + || || * + || || (1) where α and γ are the scalars which balance the strength of the low rank constraint [17] on S (the second term) and the l1 sparsity constraint [18]on S (the third term).…”

Section: Robust Dynamic Functional Connectivitymentioning

confidence: 99%

Convolutional Sparse Coded Dynamic Brain Functional Connectivity

Jin

Zhu

2018

Preprint

View full text Add to dashboard Cite

Functional brain network has been widely studied in many previous work for brain disorder diagnosis and brain network analysis. However, most previous work focus on static dynamic brain network research. Lots of recent work reveals that the brain shows dynamic activity even in resting state. Such dynamic brain functional connectivity reveals discriminative patterns for identifying many brain disorders. Current sliding window based dynamic brain connectivity framework are not easy to be applied to real clinical applications due to many issues: First, how to set up the optimal sliding window size and how to determine the threshold for the brain connectivity patterns. Secondly, how to represent the high dimensional dynamic brain connectivity pattern in a low dimensional representations for diagnosis purpose. Last, how to deal with the different length dynamic brain network patterns especially when the raw data are of different length. In order to address all those above issues, we proposed a new framework, which employs multiple scale sliding windows and automatically learns a sparse and low ran dynamic brain functional connectivity patterns from raw fMRI data. Furthermore, we are able to measure different length dynamic brain functional connectivity patterns in an equal space by learning a sparse coded convolutional filters. We have evaluated our method with state of the art dynamic brain network methods and the results demonstrated the strong potential of our methods for brain disorder diagnosis in real clinical applications.

show abstract

“…Some examples of multi-view learning algorithms include: multi-view support vector machines [18], multi-view Boosting [19], multi-view k -means [20], and clustering via canonical correlation analysis [21]. However, barring a few exceptions (e.g., multi-view feature selection methods [22], and multi-view representation learning [23]) the vast majority of existing multi-view learning algorithms are not equipped to effectively cope with the high-dimensionality of omics data [17]. Hence, predictive modeling from multi-omics data calls for effective methods for multi-view feature selection or dimensionality reduction.…”

Section: Introductionmentioning

confidence: 99%

Min-redundancy and max-relevance multi-view feature selection for predicting ovarian cancer survival using multi-omics data

et al. 2018

View full text Add to dashboard Cite

BackgroundLarge-scale collaborative precision medicine initiatives (e.g., The Cancer Genome Atlas (TCGA)) are yielding rich multi-omics data. Integrative analyses of the resulting multi-omics data, such as somatic mutation, copy number alteration (CNA), DNA methylation, miRNA, gene expression, and protein expression, offer tantalizing possibilities for realizing the promise and potential of precision medicine in cancer prevention, diagnosis, and treatment by substantially improving our understanding of underlying mechanisms as well as the discovery of novel biomarkers for different types of cancers. However, such analyses present a number of challenges, including heterogeneity, and high-dimensionality of omics data.MethodsWe propose a novel framework for multi-omics data integration using multi-view feature selection. We introduce a novel multi-view feature selection algorithm, MRMR-mv, an adaptation of the well-known Min-Redundancy and Maximum-Relevance (MRMR) single-view feature selection algorithm to the multi-view setting.ResultsWe report results of experiments using an ovarian cancer multi-omics dataset derived from the TCGA database on the task of predicting ovarian cancer survival. Our results suggest that multi-view models outperform both view-specific models (i.e., models trained and tested using a single type of omics data) and models based on two baseline data fusion methods.ConclusionsOur results demonstrate the potential of multi-view feature selection in integrative analyses and predictive modeling from multi-omics data.Electronic supplementary materialThe online version of this article (10.1186/s12920-018-0388-0) contains supplementary material, which is available to authorized users.

show abstract

MRM-Lasso: A Sparse Multiview Feature Selection Method via Low-Rank Analysis

Cited by 60 publications

References 35 publications

Prognostic Value and Potential Regulatory Mechanism of Alternative Splicing in Geriatric Breast Cancer

Prognostic Value and Potential Regulatory Mechanism of Alternative Splicing in Geriatric Breast Cancer

Convolutional Sparse Coded Dynamic Brain Functional Connectivity

Min-redundancy and max-relevance multi-view feature selection for predicting ovarian cancer survival using multi-omics data

Contact Info

Product

Resources

About