/CNNPromoterData

Data files of promoter and non-promoter sequences used to build CNN models

This directory contains bacterial and eukaryotic promoter and non-promoter sequences studied in
 Recognition of Prokaryotic and Eukaryotic Promoters Using Convolutional Deep Learning Neural Networks
      Ramzan Umarov, Victor Solovyev (PLOS One, 2016).
  Bacterial non-promoter sequences were taken from the corresponding genome sequences: we randomly selected 
fragments of protein-coding genes and took their opposite (non-coding) chain sequences. 
Escherichia coli σ70 promoter sequences were extracted from manually curated RegulonDB [1].
Bacillus subtilis promoters were taken from a collection described in [2].
  As Human, mouse and Arabidopsis non-promoter sequences (size 251 nt) we used random fragments of their genes 
located after first exons.  Eukaryotic promoter sequence regions were extracted from the well-known EPDnew database [3].
  Also additional mouse TATA promoters set was extracted from DBTSS [4]. 
[1] Gama-Castro S, Salgado H, Santos-Zavaleta A, Ledezma-Tejeida D. et al. (2016) RegulonDB version 9.0: 
high-level integration of gene regulation, coexpression, motif clustering and beyond. Nucleic Acids Res. 44(D1), D133-143.
[2] Ishii T., Yoshida K., Terai G., Fijita Y., Nakai K. (2001) DBTBS: a database of Bacillus subtilis promoters 
and transcription factors. Nucl. Acids Res., 29, 1, 278-280.
[3] 31.	Dreos, R., Ambrosini, G., Périer, R., Bucher, P. (2013) EPD and EPDnew, high-quality promoter resources in the i
next-generation sequencing era. Nucleic Acids Research, 41(Database issue): D157-64.
[4] Suzuki A, Wakaguri H, Yamashita R, Kawano S, Tsuchihara K, Sugano S, Suzuki Y, Nakai K. 
DBTSS as an integrative platform for transcriptome, epigenome and genome sequence variation 
data. Nucleic Acids Res. 2015 (Database issue) 43 (D1): D1-D5.