Network-Constrained Group Lasso for High-Dimensional Multinomial Classification with Application to Cancer Subtype Prediction
Classic multinomial logit model, commonly used in multiclass regression problem, is restricted to few predictors and does not take into account the relationship among variables. It has limited use for genomic data, where the number of genomic features far exceeds the sample size. Genomic features su...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
SAGE Publishing
2014-01-01
|
Series: | Cancer Informatics |
Online Access: | https://doi.org/10.4137/CIN.S17686 |
id |
doaj-8c178b24766e4ea0ab098c00424b0c20 |
---|---|
record_format |
Article |
spelling |
doaj-8c178b24766e4ea0ab098c00424b0c202020-11-25T02:48:37ZengSAGE PublishingCancer Informatics1176-93512014-01-0113s610.4137/CIN.S17686Network-Constrained Group Lasso for High-Dimensional Multinomial Classification with Application to Cancer Subtype PredictionXinyu Tian0Xuefeng Wang1Jun Chen2 Department of Applied Mathematics and Statistics, Stony Brook University, Stony Brook, NY, USA. Department of Preventive Medicine, Stony Brook University, Stony Brook, NY, USA. Division of Biomedical Statistics and Informatics, Mayo Clinic, Rochester, MN, USA.Classic multinomial logit model, commonly used in multiclass regression problem, is restricted to few predictors and does not take into account the relationship among variables. It has limited use for genomic data, where the number of genomic features far exceeds the sample size. Genomic features such as gene expressions are usually related by an underlying biological network. Efficient use of the network information is important to improve classification performance as well as the biological interpretability. We proposed a multinomial logit model that is capable of addressing both the high dimensionality of predictors and the underlying network information. Group lasso was used to induce model sparsity, and a network-constraint was imposed to induce the smoothness of the coefficients with respect to the underlying network structure. To deal with the non-smoothness of the objective function in optimization, we developed a proximal gradient algorithm for efficient computation. The proposed model was compared to models with no prior structure information in both simulations and a problem of cancer subtype prediction with real TCGA (the cancer genome atlas) gene expression data. The network-constrained mode outperformed the traditional ones in both cases.https://doi.org/10.4137/CIN.S17686 |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Xinyu Tian Xuefeng Wang Jun Chen |
spellingShingle |
Xinyu Tian Xuefeng Wang Jun Chen Network-Constrained Group Lasso for High-Dimensional Multinomial Classification with Application to Cancer Subtype Prediction Cancer Informatics |
author_facet |
Xinyu Tian Xuefeng Wang Jun Chen |
author_sort |
Xinyu Tian |
title |
Network-Constrained Group Lasso for High-Dimensional Multinomial Classification with Application to Cancer Subtype Prediction |
title_short |
Network-Constrained Group Lasso for High-Dimensional Multinomial Classification with Application to Cancer Subtype Prediction |
title_full |
Network-Constrained Group Lasso for High-Dimensional Multinomial Classification with Application to Cancer Subtype Prediction |
title_fullStr |
Network-Constrained Group Lasso for High-Dimensional Multinomial Classification with Application to Cancer Subtype Prediction |
title_full_unstemmed |
Network-Constrained Group Lasso for High-Dimensional Multinomial Classification with Application to Cancer Subtype Prediction |
title_sort |
network-constrained group lasso for high-dimensional multinomial classification with application to cancer subtype prediction |
publisher |
SAGE Publishing |
series |
Cancer Informatics |
issn |
1176-9351 |
publishDate |
2014-01-01 |
description |
Classic multinomial logit model, commonly used in multiclass regression problem, is restricted to few predictors and does not take into account the relationship among variables. It has limited use for genomic data, where the number of genomic features far exceeds the sample size. Genomic features such as gene expressions are usually related by an underlying biological network. Efficient use of the network information is important to improve classification performance as well as the biological interpretability. We proposed a multinomial logit model that is capable of addressing both the high dimensionality of predictors and the underlying network information. Group lasso was used to induce model sparsity, and a network-constraint was imposed to induce the smoothness of the coefficients with respect to the underlying network structure. To deal with the non-smoothness of the objective function in optimization, we developed a proximal gradient algorithm for efficient computation. The proposed model was compared to models with no prior structure information in both simulations and a problem of cancer subtype prediction with real TCGA (the cancer genome atlas) gene expression data. The network-constrained mode outperformed the traditional ones in both cases. |
url |
https://doi.org/10.4137/CIN.S17686 |
work_keys_str_mv |
AT xinyutian networkconstrainedgrouplassoforhighdimensionalmultinomialclassificationwithapplicationtocancersubtypeprediction AT xuefengwang networkconstrainedgrouplassoforhighdimensionalmultinomialclassificationwithapplicationtocancersubtypeprediction AT junchen networkconstrainedgrouplassoforhighdimensionalmultinomialclassificationwithapplicationtocancersubtypeprediction |
_version_ |
1724747561481273344 |