Hi everyone!
We ran Lexogen’s QuantSeq 3' mRNA-Seq Library Prep Kit FWD for Illumina on RNA of 49 multiplexed human samples in 4 lanes, barcoded with i7 Lexogen’s indexes, in NextSeq550. During demultiplexing with bcl2fastq, as provided by Illumina on the NextSeq550 and with the information for the sample sheet extracted from Lexogen’s i7 indexes, many reads (almost ~49% of the reads, uniformly distributed in all 4 lanes) were returned as “undetermined”, appended in 4 different fastq files (one for each lane). Has anybody else run into such a situation before? If yes, any idea why this is happening and if it is possible to demultiplex those reads, when bcl2fastq cannot demultiplex them? I have not run any code for the bcl2fastq. The software was already installed in the Nextseq550 machine and the demultiplexing was applied by the machine.
Additional data information can be provided but I am not sure what else I should provide here without any hint.
Thank you very much,
Kleio
Not an answer to your question, but It seems like you should contact Lexogen technical support as well.
This could be because of various issues. Have you looked to see what indexes are present in this pool? Do they have N's? Index issues can be due to over-loading or library problems. If there was a suspected problem with this run, was Illumina tech support consulted to see if they can diagnose the problem.
BTW: QuantSeq requires a special protocol for data analysis, in case you were not aware.
There was no mistake during the library preparation, since the bioanalyzer returned nice results. And as far as the loading is concerned, we “under-loaded”. I checked the “undetermined” indexes. During the demultiplexing, a file is produced for each lane that includes also the most Most Popular Index Sequences that were appended in the “undetermined” fastqs. Out of the 100 mostly found, only 3 have Ns and one is only Ns (NNNNNN). And those total number of reads is quite low. Thank you for the data analysis input. I am aware, but it starts with the fastqs.
If you can't correlate indexes found in the "undetermined" pool with indexes you expected then I doubt we can do anything here to help. You will need to contact Lexogen and/or Illumina tech support to see what could possibly have gone wrong. I assume you have already tried to allow for errors in indexes during the demultiplexing process to see if you can recover some of the data.
I have some code in this answer that can give you a comprehensive list of indexes found in the undetermined files, if you had not analyzed them completely: C: Demultiplexing reads with index present in the labels