Integration of Count Difference and Curve Similarity in Negative Regulatory Element Detection

Na He; Wenjing Wang; Chao Fang; Yongjian Tan; Li Li; Chunhui Hou

doi:10.3389/fgene.2022.818344

Integration of Count Difference and Curve Similarity in Negative Regulatory Element Detection

Front Genet. 2022 Feb 18:13:818344. doi: 10.3389/fgene.2022.818344. eCollection 2022.

Authors

Na He^{1

2}, Wenjing Wang³, Chao Fang⁴, Yongjian Tan², Li Li^{5

6}, Chunhui Hou²

Affiliations

¹ Harbin Institute of Technology, Harbin, China.
² Department of Biology, School of Life Sciences, Southern University of Science and Technology, Shenzhen, China.
³ School of Life Science and State Key Laboratory of Agrobiotechnology, The Chinese University of Hong Kong, Hong Kong, Hong Kong SAR, China.
⁴ Cancer Centre, Faculty of Health Sciences, University of Macau, Macao, China.
⁵ Department of Bioinformatics, Huazhong Agricultural University, Wuhan, China.
⁶ Hubei Key Laboratory of Agricultural Bioinformatics, Huazhong Agricultural University, Wuhan, China.

Abstract

Negative regulatory elements (NREs) down-regulate gene expression by inhibiting the activities of promoters or enhancers. The repressing activity of NREs can be measured globally by massively parallel reporter assays (MPRAs). However, most existing algorithms are designed for the statistical detection of positively enriched signals in MPRA datasets. To identify reduced signals in MPRA experiments, we designed a NRE identification program, fast-NR, by integrating the count and graphic features of sequenced reads to detect NREs using datasets generated by experiments of self-transcribing active regulatory region sequencing (STARR-seq). Fast-NR identified hundreds of silencers in human K562 cells that can be validated by independent methods.

Keywords: count difference; curve similarity; negative regulatory element; silencer; silencer identification.