paperKB
coga / coga-kb
Help
Sign in

Chunk #104 — Quantification and Statistical Analysis — Identifying batches in CMap data

Source
A Next Generation Connectivity Map: L1000 Platform and the First 1,000,000 Profiles.
Embedded
yes

Text

Because the CMap resource was generated over a number of years, it is inherently comprised of multiple smaller batches of data. The most predominant form of batch is the 384 well plate in which each experiment was performed. Samples on a given physical plate were processed together in the lab (i.e cell plating, treatments, amplification and detection). To mitigate plate-level effects, normalization and differential expression are computed within individual plates (see data levels 3 and 4 above). Each profile is labelled with the name of the plate and the individual well in which the experiment was done. These fields are named ‘rna_plate’ and ‘rna_well’, respectively, in the provided sample metadata.