Representation of features as images with neighborhood dependencies for compatibility with convolutional neural networks

Authors

Omid Bazgir, Texas Tech University
Ruibo Zhang, Texas Tech University
Saugato Rahman Dhruba, Texas Tech University
Raziur Rahman, Texas Tech University
Souparno Ghosh, Texas Tech University & University of Nebraska-LincolnFollow
Ranadip Pal, Texas Tech UniversityFollow

ORCID IDs

http://orcid.org/0000-0002-9311-5901

Document Type

Article

Date of this Version

2020

Citation

NATURE COMMUNICATIONS | (2020)11:4391 | https://doi.org/10.1038/s41467-020-18197-y | www.nature.com/naturecommunications

Comments

The Author(s) 2020

Abstract

Deep learning with Convolutional Neural Networks has shown great promise in image-based classification and enhancement but is often unsuitable for predictive modeling using features without spatial correlations. We present a feature representation approach termed REFINED (REpresentation of Features as Images with NEighborhood Dependencies) to arrange high-dimensional vectors in a compact image form conducible for CNN-based deep learning. We consider the similarities between features to generate a concise feature map in the form of a two-dimensional image by minimizing the pairwise distance values following a Bayesian Metric Multidimensional Scaling Approach. We hypothesize that this approach enables embedded feature extraction and, integrated with CNN-based deep learning, can boost the predictive accuracy. We illustrate the superior predictive capabilities of the proposed fra- mework as compared to state-of-the-art methodologies in drug sensitivity prediction scenarios using synthetic datasets, drug chemical descriptors as predictors from NCI60, and both transcriptomic information and drug descriptors as predictors from GDSC.