Research update – some quick tips on Illumina high throughput sequencing

August 29, 2012

For the past 3 months I’ve been working on Illumina high throughput sequencing approaches for identifying microbial communities in environmental samples using 16s rDNA.  I’ve spent the past 3 weeks troubleshooting library preparation problems and learned some interesting and possible experimental pitfalls that I thought I should share:

  • Most kits are prescribed for WGS: Most commercial kits are designed for whole genome shotgun sequencing so if you’re looking to do analysis on PCR amplicons, be prepared to Frankenstein some kits (which will require additional optimization). I wanted to multiplex 12 different samples in a lane; however, the kit only provided 10 reactions. When I asked for the next scale up, the sales rep offered to sell me a 48 reaction kit for 1,500$. In that case, you’re probably better off designing your own adapter addition strategy (be in T4 ligation or PCR)
  • Multiplexing is cheaper, but it comes at a price: If you don’t plan on normalizing your samples with qPCR or the Agilent Bioanalyzer, expect to pay a little extra for commercial labs to mix your samples. One company quoted me an additional $100 per sample in each lane. If your budget is small, you may as well normalize the DNA yourself.
  • You’ll wait longer for your data in academic labs: That’s sort-of a no brainer, but something to keep in mind. While commercial labs are marginally higher, the turnaround time is half as long and can also be a potential partnership.
  • Make sure to have enough DNA: While the sequencing reaction occurs at a pM level, if you are size selection the old school way (UV and razor blade party!) it is extremely easy to contaminate and lose yield. Most kits call for about 1 µg – give yourself a >2 µg wiggle room. Additionally, I suggest the Perkin Elmer’s LabChip device which does automated size selection for you.
  • SeqAnswers and NGS forums are your best resource: This includes troubleshooting problems, learning the technology (let’s be honest, Illumina has made it difficult for us to figure out WTF is going on), and snagging those coveted “Illumina-approved” barcoding sequences.
  • MiSeq runs are faster, slightly more expensive, and provide longer reads: Don’t automatically assume that HiSeq is the best option. The selling point with HiSeq is the 100 fold increase in coverage. It’s a bummer that the read length is stuck at 150 bp (even with the most recent upgrades) and if length is important (in this case, it is for 16s) consider using a MiSeq even though you get less reads.

Right now I am comparing and contrasting commericial adaptor ligation methods and a recently customized method published by the Knight lab. I still haven’t quite solidified the assembly pipeline, though Qiime is the analysis tool of choice.

Many thanks to Russell from the Eisen lab, the sequencing core facility staff at UCSF, and the DeRisi lab at UCSF.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: