I am trying to reproduce the preprocessing of paired-end sequencing reads in Lake et al. (ref. 1).
First, paired-end reads were filtered out if read 1 had more than four non-T bases in the last ten bases (to remove all non-poly(T)-captured contaminated reads), or had one or more bases with a poor quality score (<10). Cell barcode and UMI information were then inferred from the first 12 bases and the next 8 bases of read 1, respectively. The right mate of each read pair was trimmed to remove any portion of the SMART adaptor sequence or large stretches of poly(A) tails (6 consecutive bp or larger).
I was able to filter out reads as described in the first sentence.
However, I am unsure on which tool or procedure to use to demultiplex based on cell barcode and UMI, and to trim the right mate. Do you know what I can use or something I can read to understand how to do it?
- Lake, B. B. et al. Integrative single-cell analysis of transcriptional and epigenetic states in the human adult brain. Nature Biotechnology 36, 70–80 (2017).