Markiyan Samborskyy: Evolution of Illumina sequencing technologies and implications of reads quality and length changes for downstream data analysis.

Over the last decade there was a massive growth of the throughput of Illumina sequencing instruments, which initially was accompanied with improvement in both reads length and quality with decreasing error rates. Reads length peaked at 250 bps with average error rates at 0.05 – 0.1% achieved by MiSEQ and HiSEQ 2500 instruments in rapid run mode. Unfortunately those quality and read length trends had reversed after 2014 due to inherent limitations of the technology combined with further push to increase the raw sequence throughput and decrease the costs of manufacture.

  1. The introduction of the two channel imaging and reduction of the read length and accuracy due to decreased dephasing compensation dynamic range when compared with four channel imaging (HiSeq and MiSeq instruments). Two channel imaging is present in NextSeq, MiniSeq and NovaSeq systems and allows to halve the number of images taken (and running time) per basepair of raw output while increasing the error rate by 3-10 fold and limiting usable read length to 130-150bp. [1].

  2. The development of patterned flowcells using the ExAmp chemistry based on Recombinant Polymerase Amplification and increase of the input library quality requirements in the HiSeqX, HiSeq3000, HiSeq4000 and NovaSeq instruments. Those instruments have a much narrower input library concentration range. [2]. Also ExAmp chemistry has higher incidence of index hopping, which can affect multiplexed RNAseq and CHIPseq analysis, unless the mitigation measure in form of dual unique indexing is taken. [3,4]

[1]. A first look at Illumina’s new NextSeq 500. seqanswers. 2014-2016 http://seqanswers.com/forums/showthread.php?t=40741

[2]. (almost) everything you wanted to know about @illumina HiSeq 4000…and some stuff you didn’t. CoreGenomics. 2016/01. http://core-genomics.blogspot.co.uk/2016/01/almost-everything-you-wanted-to-know.html

[3]. Mixing sample types in a flowcell lane generates cross contamination artefacts. Simon Andrews. 2016/04. https://sequencing.qcfail.com/articles/mixing-sample-types-in-a-flowcell-lane-generates-cross-contamination-artefacts/

[4]. Effects of Index Misassignment on Multiplexing and Downstream Analysis. Illumina. 2017. https://www.illumina.com/content/dam/illumina-marketing/documents/products/whitepapers/index-hopping-white-paper-770-2017-004.pdf

0

Your Cart