Demultiplexing

[pingu]

Hello,
I have to demultiplex my fastq file, I have seen that about 200 reads are not assigned (total is 17000), there is a tool that can check barcode on 5' and 3'?
Thank you very much

[GenoMax]

Are you referring to an internal barcode that is present in the read itself? Do you know what those barcodes are?

[pingu]

Thank you for your help, I know the barcodes and they are at the of the reads (last 10bp)
For example, this read is not assigned:
@JBMBIZC01CKLPI
CACTGTAGAAGACTCGGCAGCATCTCCATGAGCCAGTATTGAAAATGTTGAAGATCAAAAAA
CACTAAGTTTTTCCAAAGTTAATATCCAATGTAAAAGATAGCAAATGCATACCCACAAACTGTA
AATGAAGATATTTGCGTTGAGGAACTTGTGACTAGCTCTTCACCCTGCAAAAATAAAAATGCA
GCCATTAAATTGTCCATATCTAATAGTAATAATTTTGAGGTAGGGCCACCTGCATTTAGGATAG
CCAGTGGTAAAATCGTTTGTGTTTCACATGAAACATTAAAAAAGTGAAAGACATATTTACAGA
CAGTTTCAGTAAAGTAATTAAGGAAAACAACGAGAATAAATCAAAAATTTGCCAAACGAAAA
TTATGGCAGGTTGTTTGGAGAACAGTGACGATCGCCTACAGTGCT

The barcodes of the reads is: AGCACTGTAG and it is a complementary reverse: CTACAGTGCT,
I have MID at the start of the reads (5') e at the end (3'), the barcode splitter tool checks MID at 5' and I have allowed 2 mismatches.

[sklages]

Looks like 454 data. If you have the original SFF file you could use Roche's sfffile tool to demultiplex your data. It recognizes 5' and 3' barcodes/MIDs.