Fasim-LongTarget enables fast and accurate genome-wide lncRNA/DNA binding prediction

Yujian Wen; Yijin Wu; Baoyan Xu; Jie Lin; Hao Zhu

doi:10.1016/j.csbj.2022.06.017

Fasim-LongTarget enables fast and accurate genome-wide lncRNA/DNA binding prediction

Comput Struct Biotechnol J. 2022 Jun 18:20:3347-3350. doi: 10.1016/j.csbj.2022.06.017. eCollection 2022.

Authors

Yujian Wen¹, Yijin Wu¹, Baoyan Xu¹, Jie Lin¹, Hao Zhu^{1

2}

Affiliations

¹ Bioinformatics Section, School of Basic Medical Sciences, Southern Medical University, Guangzhou 510515, China.
² Guangdong-Hong Kong-Macao Greater Bay Area Center for Brain Science and Brain-Inspired Intelligence, Southern Medical University, Guangzhou 510515, China.

Abstract

Many long noncoding RNAs (lncRNAs) can bind to DNA sequences proximal and distal to abundant genes, thereby regulating gene expression by recruiting epigenomic modification enzymes to binding sites. Because a lncRNA's target genes scattering in a genome have correlated functions, epigenetic analyses should often be genome-wide on both genome and transcriptome levels. Multiple tools have been developed for predicting lncRNA/DNA binding, but fast and accurate genome-wide prediction remains a challenge. Here we report Fasim-LongTarget (a revised version of LongTarget), compare its performance with TDF and LongTarget using the experimental data of the lncRNA MEG3, NEAT1, and MALAT1, and describe a case of genome-wide prediction. Fasim-LongTarget is as accurate as LongTarget and more accurate than TDF and is 200 times faster than LongTarget, making accurate genome-wide prediction feasible. The code is available on the Github website (https://github.com/LongTarget/Fasim-LongTarget), and the online service is available on the LongTarget website (https://lncRNA.smu.edu.cn).

Keywords: Epigenetic regulation; LongTarget; RNA/DNA binding; TDF; Triplex; lncRNA.