# README.txt

# Readme for CAGE clusters with npIDR'
# contact: angelika.merkel@crg.eu


# Pooled CAGE clusters ####

Pooled clusters generated by merging clusters from individual bioreplicates using mergedBed (for RNAseq experiments were two bioreplicates were available). Expression was scored individually for each replicate. npIdr was run on the normalized expression scores (E-score= number of reads mapping to a merged cluster/ number of total mapped reads).

# Format (bed6 +3):

# chromosome
# start
# end
# cluster_id = chr:start-end,strand,raw_tag_count
# bed_score = ave(E-score1,E-score2)
# strand
# E-score(tpm1) = log2(tags per 10x10exp8 mapped reads in replicate 1)
# E-score(tpm2) = log2(tags per 10x10exp8 mapped reads in replicate 2)
# npIDR 


# Non-Pooled CAGE clusters ####

For RNAseq experiments where only one bioreplicate was performed, no npIdr could be perform. E-score and tpm are derived from this sample only.

# bed_score = E-score(tpm1)
# E-score(tpm2) = .
# npIDR = .