Closed innagsc closed 6 years ago
Sorry for the extremely delayed response. Such behaviour may be observed due to the error-correction algorithms applied on the assemble step. E.g. having the information combined from two samples, MiXCR manages to eliminate more artificial diversity.
I have a library that has been sequenced on a HiSeq rapid mode (2 lanes/runs). When I ran it through the MiXCR analysis for each run individually, I got the same number of clones for each sample. However, when I merged the Fastq files and ran it through the MiXCR analysis, there were some samples where I got more or fewer clones than for the individual runs.
I will give an example: Run1- sample4 -1854 clones Run2- sample4 - 1854 clones Merged run (concatenating the fastq files for each sample) - sample4 - 1851 clones
Any ideas?
Thank you in advance!