Research Paper Volume 14, Issue 12 pp 5163—5176

Figure 1. Flowchart of the stepwise approach to screen for candidate transcription factors (TFs) and binding site SNPs. Upstream predictors of seven TFs, E2F4, EGR1, JUN, Sp1, TCF7L2, TP53, and CTNNB1, in osteoporosis [19]. Identification of genetic variants that may influence TFBS through bioinformatic sequence alignment. First, we used the data of a total of 74,861,556 variants (1,517 samples) obtained from the Taiwan BioBank database to screen for Taiwanese population-specific genetic variation. Then, through genetic alignment of GRCh37/hg19 obtained from the National Center for Biotechnology Information database, we found SNPs that may influence the binding affinity. SNPs with an MAF of <5% were excluded from the samples. Chromatin immunoprecipitation sequencing (ChIP-Seq) data obtained from the JASPAR database were used to confirm whether these genetic variants had a combination of the sites. No ChIP-Seq data were available for CTNNB1 validation, and this gene was thus excluded. Finally, we excluded results of the noncoding regions. The variation of 14 SNPs may influence transcription factor binding activity. DEG, differentially expressed gene; NGS, next-generation sequencing; SNP, single-nucleotide polymorphism; Ins/del, insertion/deletion; TFBS, TF binding site; MAF, minor allele frequency.