preprocessing at what point

grst / single_cell_data_integration

1 stars 0 forks source link

preprocessing at what point #2

Closed grst closed 6 years ago

grst commented 6 years ago

Before analyzing single cell data, the following steps have to be performed:

filtering (cells with low or exorbitantly high number of counts or genes)
removal of cell cycle effects
(doublet detection)
(per cell normalization)

Should these transformations be applied on each dataset individually, before removing batch effects or later, in the dataset already integrated?

grst commented 6 years ago

Probably before, for the following reasons:

filtering criteria depend on the platform used
(might) make the life of the batch removal tool easier if cell cycle effects are removed already.

grst commented 6 years ago

According to https://github.com/brianhie/scanorama/issues/15:

Scanorama correction should come late (or even last) in the data processing pipeline.

grst commented 6 years ago

RTFD

Scanorama is designed to be used in scRNA-seq pipelines downstream of noise-reduction methods, including those for normalization, imputation, and highly-variable gene filtering. The results from Scanorama integration and batch correction can then be used as input to other tools for scRNA-seq clustering, visualization, and analysis.