RT Journal Article
SR Electronic
T1 An exact transformation of convolutional kernels applied directly to DNA/RNA sequences
JF bioRxiv
FD Cold Spring Harbor Laboratory
SP 163220
DO 10.1101/163220
A1 Yang Ding
A1 Jingyi Li
A1 Meng Wang
A1 Ge Gao
YR 2017
UL http://biorxiv.org/content/early/2017/07/13/163220.abstract
AB Motivation The powerful learning ability of a convolutional neural network (CNN) to perform functional classification of DNA/RNA sequences could provide valuable clues for the discovery of underlying biological mechanisms. Currently, however, the only way to interpret the direct application of a convolutional kernel to DNA/RNA sequences is the heuristic construction of a position weight matrix (PWM) from fragments scored highly by that kernel; whether the resulting PWM still performs the sequence classification well is unclear.Results We developed a novel kernel-to-PWM transformation whose result is theoretically provable. Specifically, we proved that the log-likelihood of the resulting PWM of any DNA/RNA sequence is exactly the sum of a constant and the convolution of the original kernel on the same sequence. Importantly, we further proved that the resulting PWM demonstrates the same performance, in theory, as the original kernel under popular CNN frameworks. Surprisingly, our PWMs almost always outperformed heuristic ones at sequence classification, whether the discriminative motif was sequenceor structure-conserved. These results compelled us to further develop a maximum likelihood estimation of the optimal PWM for each kernel and a back-transformation of predefined PWMs into kernels. These tools can benefit the biological interpretation of kernel signals.Availability Python scripts for the transformation from kernel to PWM, the inverted transformation from PWM to kernel, and the maximum likelihood estimation of optimal PWM are available through ftp://ftp.cbi.pku.edu.cn/pub/software/CBI/k2p.Contact gaog{at}mail.cbi.pku.edu.cn Supplementary information Supplementary data are available at Bioinformatics online.