I disagree - choice of kit matters significantly.

I recently did a test comparing Agilent, Nimblegen and Nextera std exome kits on a set of 63 gene relevant to our research, as well as over the whole exome. The coverage and calling ability differed between the kits and differed whether I was looking at 63 genes or the whole exome.

Probe design and coverage is different between kits and this influences your results greatly.

Get the BED files from Agilent (you need to login to get the information https://earray.chem.agilent.com/suredesign) and Nimblegen (http://www.nimblegen.com/downloads/annotation/ez_exome_v3/SeqCapEZ_Exome_v3.0_Design_Annotation_files.zip) and look at the probe coverage for your 50 genes. If the coverage looks good then stick with the std kits - otherwise custom capture is your best bet. However, be warned custom capture kits are very expensive.

If you have a small number of samples (

I disagree - choice of kit matters significantly.

I recently did a test comparing Agilent, Nimblegen and Nextera std exome kits on a set of 63 gene relevant to our research, as well as over the whole exome. The coverage and calling ability differed between the kits and differed whether I was looking at 63 genes or the whole exome.

Probe design and coverage is different between kits and this influences your results greatly.

Get the BED files from Agilent (you need to login to get the information https://earray.chem.agilent.com/suredesign) and Nimblegen (http://www.nimblegen.com/downloads/annotation/ez_exome_v3/SeqCapEZ_Exome_v3.0_Design_Annotation_files.zip) and look at the probe coverage for your 50 genes. If the coverage looks good then stick with the std kits - otherwise custom capture is your best bet. However, be warned custom capture kits are very expensive.

If you have a small number of samples (

Thanks for your valuable suggestions - Rohan and Christian.

All the best christian for future publication.

As I am dealing with rare disease, So Sample size would be small, at max. 50 Samples and targeting 50-55 genes of disease pathways and looking of rare variants and novel variants in case samples.

I have heard from many places that coverage is not good in Exome capture std kits. Is it True??

Before starting the project, I want to be sure that those 50 genes are captured with good coverage.

Which custom/std kit is best?? - Nimblegen, Agilent or Nextera??

Thanks

I would say that coverage is actually pretty good, on average. However, on individual genes it varies between kits: Agilent will be better on some genes and Nimblegen will be better on others. That's why you need to check *your* genes specifically.

We went with Nimblegen as we found that worked best for us, but it may not be necessarily the case for you.

Thank you very much for your reply, Sorry if i unable to make myself clear about my research question. I want to do Targeted custom enrichment for 50 genes of cardiomyopathy covering all exons, introns, UTRs, regulatory regions to find the rare and novel variants.

I want to know that what are the risk involved in custom targeted NGS experimentally and also need help from those who have performed these kind of experiments with custom targeted NGS. What are the do's and don't of custom Targeted NGS.??

Please help.

Hello - there are several custom target-enrichment platforms out there - the two I've used on an Illumina platform are SureSelect and Haloplex, both from Agilent. SureSelect is a bit more expensive and the workflow is more intensive, but the results are excellent - I used it for an 80 gene screen for a rare childhood aplasia (DBA). Haloplex is easier and cheaper, but the quality of the data wasn't quite as good (more drop-outs, less coverage of non-coding regions, etc). They both use the same web-based design portal, and it really is as easy as entering in your genes of interest (or regions if you want non-coding as well, although exon is also available) and you're away!

Agilent Haloplex combines multiple restriction enzymes before capture, allowing to capture better some difficult regions as compared to the hybridization or amplicon only based approaches . In any case for a significant clinical project you should test yourself on the actual genes of interest and evaluate coverage and variant calling ability on a set of positive cases (i.e. where mutation is known). Also don't forget that in terms of clinical utility you might find that there are certain types of mutations, common for the disease investigated, which might not be seen well by capture NGS and might need to be added by hand.

@Rohan - the kits you used are both from Illumina (TruSeq and Nextera) and our in hands the techinical reprodicbility of the Nextera kit was much worse that for NimbleGen or Agilent (~85% vs >95%), so that's where I would expect your problems to come from. In general mRNA sequencing is very reproducible, we can get correlations of up to 0.99 between biological replicates.

There are other studies published comparing WES kits, but from a couple of years ago now. Most are in this special issue in Genome Biology http://genomebiology.com/content/12/9, plus there's one from Nature Biotech. http://www.nature.com/nbt/journal/v29/n10/full/nbt.1975.html

@Amitabh Biswas - As you are aiming at rare diseases and cardiogenetics you could go for the Agilent Haloplex 180 ISCA/ICG kit, containing 180 rare diseases relevant genes.

For cardiogenetics, the Dutch DNA-Diagnostic Labs from the Clinical Genetics departments have agreed on a sureselect design (some with a few local additions). We run these panels routinely. Dr. Slegtenhorst is the lead CMG for that cohort. One or two labs use their exome-kit and fill up the gaps by Sanger-seq.

On some cut-off point your lab will have to decide whether the test will follow Sanger-quality (and fill up drops and missing exons by PCR) or offer a screening-quality (still with a good yield though!)