Table 1.
Overview of the data sets properties used in this study 
: From left to right: data sets source, in vivo next-generation sequencing technique, number of positive and negative labels and the partition of TSS shared with at least one other data sets. The annotations are derived or mapped (Cappable-seq) on E. coli MG1655 (accession: ‘NC_000913.3’). The total genome consists of 9 283 304 nucleotides.
| Data source | Technique | Year | Positive labels | Negative labels | Shared TSSs |
|---|---|---|---|---|---|
| RegulonDB [29] | Collection | – | 6487 (0.07%) | 9 276 817 (99.93%) | 2476 (38.17%) |
Etwiller   . [9] |
Cappable-seq | 2016 | 16 348 (0.18%) | 9 266 956 (99.82%) | 4631 (28.33%) |
Yan   . [40] |
SMRT-Cappable-seq | 2018 | 2574 (0.03%) | 9 280 730 (99.97%) | 2311 (89.78%) |
Ju   . [18] |
SEnd-seq | 2019 | 5502 (0.06%) | 9 277 802 (99.94%) | 4026 (73.17%) |