Overcoming Interpretability in Deep Learning Cancer Classification
Authors:
- 20 February 2021
- book chapter
- Published by Springer Nature
- Vol. 2243, 297-309
Abstract
Since its inception, deep learning has revolutionized the field of machine learning and data-driven science. One such data-driven science to be transformed by deep learning is genomics. In the past decade, numerous genomics studies have adopted deep learning and its applications range from predicting regulatory elements to cancer classification. Despite its dominating efficacy in these applications, deep learning is not without drawbacks. A prominent shortcoming of deep learning is the lack of interpretability. Hence, the main objective of this study is to address this obstacle in the deep learning cancer classification. Here we adopt a feature importance scoring methodology (Gradient-based class activation mapping or Grad-CAM) on a quasi-recurrent neural network model that classify cancer based on FASTA sequencing data. In this study, we managed to formulate a nucleotide-to-genomic-region Grad-CAM scoring methodology, as well as, validate the use this methodology for the chosen model. Consequently, this allows for the utilization of the Grad-CAM scoring methodology for feature importance in deep learning cancer classification. The results from our study identify potential novel candidate genes, genomic elements, and mechanisms for future cancer research.Keywords
This publication has 26 references indexed in Scilit:
- Deep learning: new computational modelling techniques for genomicsNature Reviews Genetics, 2019
- Sequential regulatory activity prediction across chromosomes with convolutional neural networksGenome Research, 2018
- DeepCpG: accurate prediction of single-cell DNA methylation states using deep learningGenome Biology, 2017
- Predicting the impact of non-coding variants on DNA methylationNucleic Acids Research, 2017
- Basset: learning the regulatory code of the accessible genome with deep convolutional neural networksGenome Research, 2016
- Predicting effects of noncoding variants with deep learning–based sequence modelNature Methods, 2015
- Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research GroupsIEEE Signal Processing Magazine, 2012
- Multi-column deep neural networks for image classification2012
- Developments and directions in speech recognition and understanding, Part 1 [DSP Education]IEEE Signal Processing Magazine, 2009
- HYBRID NEURAL NETWORK/HIDDEN MARKOV MODEL SYSTEMS FOR CONTINUOUS SPEECH RECOGNITIONInternational Journal of Pattern Recognition and Artificial Intelligence, 1993