What to do with leading T's and G's?

[dschika]

Hello all,

I have several cDNA libraries sequenced with 454 Titanium. My problem is that almost all sequences in the libraries (after clipping MID) start with a sequence of 'TG' mostly 'TGTGTTGGGTGTGTTTGG'. But the length and number of T's and G's can also vary. I asked the provider what this is and what to do with that and they told me that those sequences are relicts of the isothermal amplification and that I could simply cut all T's and G's until I reach the first A or C…Has anybody made experiences with that or heard about that and can tell me if it is really ok to just trim the leading T's and G's?

Thanks a lot!

Best regards

[0Gen]

Since you mentioned isothermal amplification, what method did you use for sample amplification?

I had the same problem - TGs on one side and ACs at the tail (if the tail was not trimmed). My samples were amplificed by Genoplex (Sigma kit). So, TGs and ACs are the two primers used in Genoplex, lengths of TGs and ACs are varied though.

[Dethecor]

Clipping Nucleotides that are artifacts of the sequencing process

Hi,

I had a similar situation when we used a process for the reverse transcription (in RNA-Seq) that would add a couple of G's at the 3'-End of some of the Reads.

For me simply clipping all leading G's was not a feasible solution since I needed spatial information about where exactly the reads started.

If you can safely remove those leading nucleotides greatly depends on what you want to do with the library afterwards, because it obviously influences the nucleotide distribution at each position of your sequences etc.

I think you just have to keep in mind that you will cut away leading T's and G's that actually carry information that should be represented in the library (because you can't distinguish between those and the artifacts from the isothermal amplification).

Basically you are just introducing another bias that you might have to compensate for in subsequent analysis, because all your reads will start either with an A or a C.

- Cheers

[dschika]

Since you mentioned isothermal amplification, what method did you use for sample amplification?

I had the same problem - TGs on one side and ACs at the tail (if the tail was not trimmed). My samples were amplificed by Genoplex (Sigma kit). So, TGs and ACs are the two primers used in Genoplex, lengths of TGs and ACs are varied though.

Yes, the person of the provider to I talked to mentioned the Sigma kit but wasn't sure if the TGs and the ACs (which I have also found) are primers...good to know

In one of the libraries, which had to be sequenced a second time, the sequences start with something like 'CTCGCGTGTCTGTGTTGGTG' - so i guess this is just a slightly different primer or primer of another amplification kit?

I think you just have to keep in mind that you will cut away leading T's and G's that actually carry information that should be represented in the library (because you can't distinguish between those and the artifacts from the isothermal amplification).

Basically you are just introducing another bias that you might have to compensate for in subsequent analysis, because all your reads will start either with an A or a C.

You are definitively right. That's why I have a "bad feeling" about simply clipping leading TGs and ending ACs...but I don't see another possibility to get rid of those parts at the moment . Perhaps I also clip the first found A or C. More information which is lost but probably not all sequences will start with an A or a C.

Thanks for your replies!