.bcl to *qseq.txt conversion

[E_Klee]

We are just starting to look at the informatics needs for supporting a HiSeq 2000. Does anyone have estimates for time and processing requirements to do the .bcl to *qseq.txt file format conversion?

Thanks

[csquared]

Less than an hour per paired-100 flowcell on a reasonably fast cluster. We put each flowcell on a single node (8 cores) and it finishes in about 40 minutes.

[E_Klee]

Thanks

Thanks for the time estimates. Do you also have any estimates as to how long it takes to demultiplex an 8 sample (or 12), per lane, run?

[csquared]

Demultiplexing is pretty fast but depends on what barcoding method is being used.

Whose barcodes are you using? If it is Illumina's indexing using the 3 reads, have you used them in the past? If not, now may not be the best time to start. I would wait a couple months and then re-evaluate barcode methods. If it is the adaptor-based barcodes, demultiplexing a 7bp barcode using typical Hamming error correction takes about 30 minutes for a 12-plex.