Recent publications have shown how far we have come. First sequencing of circulating tumour DNA (Forshew et al), then sequencing of a Human foetus (Kitzman et al), MALBAC single-cell CNV (Zong et al), cell-free DNA cancer genome sequencing (Allen Chan et al) and now ctDNA exome sequencing of tumours (Murtaza et al). The first three are very briefly reviewed at the bottom of this post and you can read all about ctExome-seq in my next post!
Much of the advance has come about through technological improvements, the NGS itself being the most obvious, but in the papers the sample prep is often just as significant a technological leap. Forshew et al used Fluidigm’s Access Array and I have posted about that previously. The papers from Jay Shendure’s and Nitzan Rosenfeld's groups (Kitzmand and Murtaza) both used Rubicon Genomics’ Thruplex kits to prepare samples from circulating DNA.
Rubicon Genomics Thruplex: The ThruPLEX-FD kit uses a proprietary set of end-repair, ligation, and amplification reactions to make Illumina sequencing libraries from 50pg to 50ng of input DNA. It is fast and simple to process, requiring a couple of hours in the lab. All reactions are carried out in a single tube 3 step protocol and up to 12 samples can be indexed for multiplexed sequencing. It can be used for genomic, plasma, FFPE, cDNA and ChIP DNA. Rubicon claim a 50-100X increased in sensitivity over Illumina TruSeq or NEBnext library preps. They get this increase by using a proprietary high-efficiency ligation and optimized reagents to maximise input DNA to sequencing library conversion with a very low background.
The Thruplex Protocol “simples":
- Mix DNA with template preparation reagents and incubate at 22C, then 55C for 45 minutes.
- Add library synthesis reagents and incubate at 22C for 40 minutes.
- PCR amplify with library amplification reagents for 4-21 cycles.
- Clean, pool and sequence.
It looks like Thruplex is going to offer us the ability to assay samples we previously would not have considered. As others develop competintg technologies we are likely to get to lower and lower amounts of DNA being used for standard sequencing rather than these being niche products. Long gone are the days of 1-10ug being the standard input.
Does anyone need some left-over nebulisers?
Recent “genome amplification” papers:
May 2012 Forshew, T et al: TAM-seq sequencing of circulating tumour DNA, see the blog post for more details. See Noninvasive Identification and Monitoring of Cancer Mutations by Targeted Deep Sequencing of Plasma DNA.
June 2012 Kitzman, JO et al: Thruplex sequencing of a Human foetus. Kitzman and colleagues present a tour de force of prenatal NGS. Using cell free DNA in maternal blood and comparison to the maternal and paternal genomes they were able to reconstruct a high-quality foetal genome and also discover de novo mutations on the foetal genome. This shows what may be possible when using NGS for pre-natal screening, although sequencing of specific regions is likely to be faster and more cost-effective for a while yet. See Noninvasive Whole-Genome Sequencing of a Human Fetus.
Dec 2012 Zong, C et al: MALBAC sequencing of single cells for SNV and CNV. The MALBAC (multiple annealing and looping-based amplification cycles) method allows single cells to be reliably analysed for CNVs. In the paper Zong and colleagues compared single cell CNVs to bulk DNA (see Figure 3 from their paper). Genome-WideDetection of Single-Nucleotide and Copy-Number Variations of a Single HumanCell.
2013 Chan, KC et al: Cancer genome sequencing from cell-free DNA. Comparison of CNV’s in the tumour to pre- and post-surgery plasma DNA sequencing in all cases tumor and pre-surgery plasma showed the same copy number abberations and these disappeared in all post-surgery plasma samples. Below is one of the circus plots from figure 1 of their paper. See Cancer genomescanning in plasma: detection of tumor-associated copy number aberrations,single-nucleotide variants, and tumoral heterogeneity by massively parallelsequencing.