Solexa reads contamination

[l.miozzi]

Hallo to all,

analysing the reads obtained from a Solexa sequencing of a library of small RNAs I obtained a sequence overrepresented (according to fastqc);
this is a sequence of 26 nts matching to the luciferase for the first 23nt, and representing more than 7 milions of reads. The luciferase is not expected to be present in my biological samples so I wonder if there is a step in the sequencing justifying this aspect.

Thank you very much for any help!

Laura

[simonandrews]

Nothing in the normal sequencing protocol uses luciferase so I doubt it's coming from the library construction. It would be much more likely to be a contamination in your starting material. However you'd not normally expect to see exactly the same sequence turn up in this case, since any contamination should be randomly fragmented during sonnication.

Is the overrepresented sequence part of the actual luciferase sequence, or could it be part of a plasmid where the top hit comes from a luciferase containing derivative? Did you get other overrepresented sequences as well? If so, can you try to assemble them to see if you can put together a longer sequence which might help clarify where it came from.

[pmiguel]

One common source of contamination of genomic DNA prep comes from re-usable centrifuge bottles. If the same bottles are being used to isolate large plasmid preps and genomic DNA, you can pick up some of the plasmid sequence if your downstream assay is sensitive enough and heroic measures (including an acid wash) are not taken to remove the high molar concentrations of plasmid from the bottles prior to the genomic DNA prep.

Similarly I have seen one case where reagents from a manufacturer were contaminated with just about every vector they ever offered for sale. We were using the their reagents quite a bit outside their design specifications, and the results were not completely disastrous.

--
Phillip