Agilent SureSelect XT Capture vs. SureSelect XT2 Capture ? What's the difference ?

[medalofhonour]

I am trying to understand the difference between Agilent's Sureselect XT vs. XT2 exon captures.

From what I understand, with XT2, you pool all of the samples together and then use the capture. (V4 Exome + UTR) while with XT, you use the capture individually on each sample.

Is this understanding correct ?

Sorry if this question is too naive, I am mainly a bioinformatics person trying to understand the process.

From a Bioinformatics point of view, do you know if Agilent has separate BED files of targeted regions for XT and XT2 ? This is because I recently analyzed a set of samples, some of which were prepped using XT and others using XT2.

And for the ones with XT2, we did not find many reads mapping in the regions defined by Agilent's V4 + UTR BED file ? But there were a good amount of reads mapping in these regions for the XT samples.

Has anyone come across something like this before ?

[Heisman]

At least a year ago when I did targeted capture I was able to access the bed files of targeted regions through earray; is that possible now? I'd give that a try.

[medalofhonour]

Yes I have the BED file, but it's just one bed file for V4 Exome + UTR capture. I am wondering if there are separate BED files for XT and XT2 protocols ?

This is because the BED file I am using gives me very low number of reads mapping in the target regions for the XT2 data.

[Cofactor Genomics]

Have you run any qc on the data (FASTQC etc)? I would be suspect of the data and not the BED file if you are seeing few reads map back to the regions. I do not believe there are different BED files for pre and post-pooling captures.

Have you taken the data and aligned to the reference genome, not the BED file, to see the percentage of reads that hit the genome? This will tell you at least if it is the right sample or if there is some other contamination going on. Something could have gotten mixed up before it came to you.

There are quite a few other simply QC procedures that can be performed as it sounds like to me that it is a data problem and not an incorrect reference.

Jon Armstrong
Cofactor Genomics

[sprabhu]

Hello,
Did anyone figure out whether this was a bed file associated issue or whether it was the pre-capture (XT2) library prep steps that affected capture efficiency?
Any updates on XT2 performance would be appreciated.
Thanks