The species information of the sequencing sequence is incorrect

Question

[1437310205]

I used HepG2 cells for RNA-SEQ, but the mapping rate was very low, only 3%. After species comparison, I found that rats accounted for the largest proportion. I don't know what went wrong. Have you ever encountered a similar situation？

[Michael]

https://en.wikipedia.org/wiki/Chinese_hamster_ovary_cell

Did you mix up your cell lines or contaminate them?

Contamination is a major problem in cell culture. This would be unfortunate, the only good thing here is that the sequencer doesn't "lie" or makes up sequences and you might therefore have noticed the problem.

• Assemble your RNAseq data with Trinity
• Extract SSU and LSU (ribosomal RNA) transcripts
• Blast theses to find which species it really is

[cpad0112]

Another way is to index possible contaminant genomes, sample reads (million or 0.1 million read) and run fastqscreen.

[Michael]

In fact, Chinese hamster is C. griseus, whereas the main hit is to gold hamster (not a rat, btw) so my initial assumption does not hold anymore, the genome was already in there. It is difficult to guess what went wrong without further information. I would expect that if the real species was really present in the reference genomes tried, one should see >70% aligned.

The problem with contaminants is that they might not always have a reference genome. But there are many more SSU LSU sequences in databases. So yes, a thorough "post mortem" analysis is required using multiple tools and setups, even though you may have to discard the data in the end anyway.