COLOMBOS (https://doi.org/10.1093/nar/gkv1251) and Ecomics (https://dx.doi.org/10.1038%2Fncomms13090) are two datasets that integrate public gene expression data using two different approaches. COLOMBOS integration strategy is based on fold-changes (in the form of log-ratios), where each gene measurement does not represent the expression value in a single sample or condition, but a change in gene expression comparing a test condition to a reference condition (a contrast), thus is a relative measurement. The Ecomics integration strategy instead, employs a normalization method to represent expression levels in absolute-scale.
The following collection of Jupyter notebooks tries to add some further insights on the peculiarities of the two distinct approaches. Both datasets are publicly available and thus all the following analysis should be easily reproducible.
For convenience, a copy of the datasets can be found at this link https://drive.google.com/open?id=1OucKsNxKiGwxtN_e8OaZys3rMkPOwXys
COLOMBOS: http://colombos.fmach.it/cws_data/compendium_data/ecoli_compendium_data.zip
Ecomics: https://www.dropbox.com/sh/lqzyd6dzmg1a2c4/AADHqUbNXzKyya_tNQOHN__Wa?dl=0
Ecomics meta-data: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5059772/bin/ncomms13090-s2.xlsx (sheet Transcriptome)