SIMLR: A Tool for Large‐Scale Genomic Analyses by Multi‐Kernel Learning

Wang, Bo; Ramazzotti, Daniele; Sano, Luca De; Zhu, Junjie; Pierson, Emma; Batzoglou, Serafim

doi:10.1002/pmic.201700232

Cited by 90 publications

(62 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Concordance index↑ Hazard ratio↑ P value CC [8] 0.531±0.046 2.90 (0.75-14.45) 0.113 SIMLR [12] 0.582±0.047 3.20 (0.73-13.93) 0.122 AE+GMM+MML 0.623±0.052 3.32 (1.35-8.18) 0.009* Table 1. Comparison of unsupervised learning algorithms for imaging phenotype clustering.…”

Section: Methodsmentioning

confidence: 99%

“…Evaluation The resulting clusters from our pipeline and their clinical values were demonstrated using a kaplan-meier plot and results from a log-rank test. We also compared our approach with other unsupervised clustering methods that have been used to cluster radiomic features, including baseline consensus clustering [8,13] and state-of-the-art SIMLR [12]. Further, we compared our approach to validated clinical and imaging biomarkers for prognostic ability [2,4].…”

Section: Experimental Designmentioning

confidence: 99%

See 1 more Smart Citation

Unsupervised Clustering of Quantitative Imaging Phenotypes Using Autoencoder and Gaussian Mixture Model

Chen

Milot

Cheung

et al. 2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Quantitative medical image computing (radiomics) has been widely applied to build prediction models from medical images. However, overfitting is a significant issue in conventional radiomics, where a large number of radiomic features are directly used to train and test models that predict genotypes or clinical outcomes. In order to tackle this problem, we propose an unsupervised learning pipeline composed of an autoencoder for representation learning of radiomic features and a Gaussian mixture model based on minimum message length criterion for clustering. By incorporating probabilistic modeling, disease heterogeneity has been taken into account. The performance of the proposed pipeline was evaluated on an institutional MRI cohort of 108 patients with colorectal cancer liver metastases. Our approach is capable of automatically selecting the optimal number of clusters and assigns patients into clusters (imaging subtypes) with significantly different survival rates. Our method outperforms other unsupervised clustering methods that have been used for radiomics analysis and has comparable performance to a state-of-the-art imaging biomarker.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Experimental Designmentioning

confidence: 99%

Unsupervised Clustering of Quantitative Imaging Phenotypes Using Autoencoder and Gaussian Mixture Model

Chen

Milot

Cheung

et al. 2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…scClassify uses a modified version of the SIMLR algorithm [40] to cluster unassigned cells. To illustrate scClassify's capacity to annotate cells that are not in the reference dataset, we trained scClassify on a reference dataset that had only four cell types (Xin et al), and used this to predict cell types for a dataset with nine cell types from human pancreas (Muraro et al).…”

Section: Component 4: Post-hoc Clustering Of Unassigned Cellsmentioning

confidence: 99%

scClassify: hierarchical classification of cells

Lin

Cao

Kim

et al. 2019

Preprint

View full text Add to dashboard Cite

17Cell type identification is a key computational challenge in single-cell RNA-sequencing 18 (scRNA-seq) data. To capitalize on the large collections of well-annotated scRNA-seq datasets, 19 we present scClassify, a hierarchical classification framework based on ensemble learning. 20 scClassify can identify cells from published scRNA-seq datasets more accurately and more 21 finely than in the original publications. We also estimate the cell number needed for accurate 22 classification anywhere in a cell type hierarchy. 23 24Single cell RNA-sequencing (scRNA-seq) datasets have become larger and more diverse, driving the 25 need for classification methods that can provide more refined discrimination between cell types. 26Many major cell types can be divided into subtypes in a hierarchical fashion, forming what we call a 27 'cell type hierarchy' 1,2 . Approaches to scRNA-seq studies that account for cell type hierarchies in 28 experimental design and cell type identification will permit more nuanced interpretations of the 29 resulting data. 3 30 31The most common approach to identifying cell types within scRNA-seq data is unsupervised 32 clustering 4-6 followed by manual annotation using known marker genes. However, the number of 33

show abstract

“…The Matlab code for CIMLR is available at https://github.com/danro9685/CIMLR; the R implementation of the tool is also available on Github (https://github.com/BatzoglouLabSU/SIMLR) and as part of the Bioconductor release of SIMLR 40 .…”

Section: Author Contributionsmentioning

confidence: 99%

Multi-omic tumor data reveal diversity of molecular mechanisms that correlate with survival

Ramazzotti

Wang

et al. 2018

Preprint

Self Cite

View full text Add to dashboard Cite

Outcomes for cancer patients vary greatly even within the same tumor type, and characterization of molecular subtypes of cancer holds important promise for improving prognosis and personalized treatment. This promise has motivated recent efforts to produce large amounts of multidimensional genomic ('multi-omic') data, but current algorithms still face challenges in the integrated analysis of such data . Here we present Cancer Integration via Multikernel Learning (CIMLR; based on an algorithm originally developed for analysis of single-cell RNA-Seq data), a new cancer subtyping method that integrates multi-omic data to reveal molecular subtypes of cancer. We apply CIMLR to multi-omic data from 32 cancer types and show significant improvements in both computational efficiency and ability to extract biologically meaningful cancer subtypes. The discovered subtypes exhibit significant differences in patient survival for 21 of the 32 studied cancer types. Our analysis reveals integrated patterns of gene expression, methylation, point mutations and copy number peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.The copyright holder for this preprint (which was not . http://dx.doi.org/10.1101/267245 doi: bioRxiv preprint first posted online Feb. 16, 2018; changes in multiple cancers and highlights patterns specifically associated with poor patient outcomes. IntroductionCancer is a heterogeneous disease that evolves through many pathways, involving changes in the activity of multiple oncogenes and tumor suppressor genes. The basis for such changes is the vast number and diversity of somatic alterations that produce complex molecular and cellular phenotypes, ultimately influencing each individual tumor's behavior and response to treatment. Due to the diversity of mutations and molecular mechanisms, outcomes vary greatly and it is therefore important to identify cancer subtypes based on common molecular features, and then correlate those with outcomes. This will lead to an improved understanding of the pathways by which cancer commonly evolves, as well as better prognosis and personalized treatment.Efforts to distinguish subtypes are complicated by the many kinds of genomic changes that contribute to cancer -for example, point mutations, DNA copy number aberrations, DNA methylation, gene expression, protein levels, and post-translational modifications. While gene expression clustering has often been used to discover subtypes (e.g., the PAM50 subtypes 1 of breast cancer), analysis of a single data type does not typically capture the full complexity of a tumor genome and its molecular phenotypes.For example, a copy number change may be biologically relevant only if it causes a gene expression change; gene expression data alone ignores point mutations that may alter the function of the gene product; and point mutations in two different genes may have the same downstream effect, which may become apparent only when also considering methylation or gene expression. Therefore, comprehen...

show abstract

SIMLR: A Tool for Large‐Scale Genomic Analyses by Multi‐Kernel Learning

Cited by 90 publications

References 12 publications

Unsupervised Clustering of Quantitative Imaging Phenotypes Using Autoencoder and Gaussian Mixture Model

Unsupervised Clustering of Quantitative Imaging Phenotypes Using Autoencoder and Gaussian Mixture Model

scClassify: hierarchical classification of cells

Multi-omic tumor data reveal diversity of molecular mechanisms that correlate with survival

Contact Info

Product

Resources

About