Adapterama V: Primers and protocols for 480 unique dual-indexed or 230,400 combinatorially-indexed Illumina bead-linked transposase libraries (iNextEra)

Efficient, low-cost conversion of DNA into a form that can be sequenced on massively parallel sequencers is critical in a variety of contexts. Two recent approaches for Illumina libraries focus on either a highly robust, but relatively expensive, library preparation using commercial bead-linked tagmentation for Nextera-style DNA libraries or a modification of that approach for low-cost libraries appropriate for low-coverage genome skimming. We modified the low-cost approach to construct libraries for capture enrichment sequencing (CES) or modest-coverage whole-genome sequencing (WGS). We optimized an intermediate amount of input DNA template (targeting 25ng, but ranging from 1-100 ng), bead-linked transposase (0.5microL ; 1/20th of the Illumina protocol), and number of PCR cycles to achieve libraries of sufficient quantity and diversity for CES and consistent genome coverage. To facilitate pooling many samples, we designed and synthesized 480 iNextEra5 and 480 iNextEra7 primers which can be used to produce 480 unique dual-indexed libraries or 230,400 combinatorially-indexed libraries. All primers include 10 nucleotide (nt) indexes that maintain edit distance more than 3, which allows error-correction. To facilitate pooling of the new (iNextEra) libraries with our previous [iTru (TruSeq-style)] libraries, we identify groups of iNextEra indexes that achieve an edit distance more than 2 with all iTru indexes. We then tested a broad range of input DNA quantities, and DNA sources (bacteria, yeast, mammalian, negative control), which was used for CES and/or WGS.