_I'm breaking up the files from jaclyn-taroni/qc_viz into smaller PRs. (That branch will be retained for posterity.)_
In 05-pca_test_compendium, we found that there are what look to be two groups of RNA-seq samples (separated in PC1). Our initial suspicion was that this had to do with the selection strategy (e.g., poly-A enrichment vs. ribo-depletion), but that didn't appear to be the case after taking a look at the methods for a handful of samples. In this notebook, I explored whether things like paired- vs. single-end, the length of the reads, the inferred library type, etc. could account for this difference. It seems that one group is comprised of large experiments (100-1000s of samples) from the Wellcome Sanger Institute Zebrafish Mutation Project, which tend to have lower mapping rates compared to a random selection of samples from the other group.
_I'm breaking up the files from
jaclyn-taroni/qc_vizinto smaller PRs. (That branch will be retained for posterity.)_In
05-pca_test_compendium, we found that there are what look to be two groups of RNA-seq samples (separated in PC1). Our initial suspicion was that this had to do with the selection strategy (e.g., poly-A enrichment vs. ribo-depletion), but that didn't appear to be the case after taking a look at the methods for a handful of samples. In this notebook, I explored whether things like paired- vs. single-end, the length of the reads, the inferred library type, etc. could account for this difference. It seems that one group is comprised of large experiments (100-1000s of samples) from the Wellcome Sanger Institute Zebrafish Mutation Project, which tend to have lower mapping rates compared to a random selection of samples from the other group.