Bead-linked transposomes enable a normalization-free workflow for NGS library preparation

Abstract

Background: Transposome-based technologies have enabled the streamlined production of sequencer-ready DNA libraries; however, current methods are highly sensitive to the amount and quality of input nucleic acid.

Results: We describe a new library preparation technology (Nextera DNA Flex) that utilizes a known concentration of transposomes conjugated directly to beads to bind a fixed amount of DNA, and enables direct input of blood and saliva using an integrated extraction protocol. We further report results from libraries generated outside the standard parameters of the workflow, highlighting novel applications for Nextera DNA Flex, including human genome builds and variant calling from below 1 ng DNA input, customization of insert size, and preparation of libraries from short fragments and severely degraded FFPE samples. Using this bead-linked library preparation method, library yield saturation was observed at an input amount of 100 ng. Preparation of libraries from a range of species with varying GC levels demonstrated uniform coverage of small genomes. For large and complex genomes, coverage across the genome, including difficult regions, was improved compared with other library preparation methods. Libraries were successfully generated from amplicons of varying sizes (from 50 bp to 11 kb), however, a decrease in efficiency was observed for amplicons smaller than 250 bp. This library preparation method was also compatible with poor-quality DNA samples, with sequenceable libraries prepared from formalin-fixed paraffin-embedded samples with varying levels of degradation.

Conclusions: In contrast to solution-based library preparation, this bead-based technology produces a normalized, sequencing-ready library for a wide range of DNA input types and amounts, largely obviating the need for DNA quantitation. The robustness of this bead-based library preparation kit and flexibility of input DNA facilitates application across a wide range of fields.

Keywords: Library preparation; Next-generation sequencing; Transposome.

Fig. 1

Bead-based tagmentation reduces insert size variation and normalizes yield. Libraries were prepared from human gDNA (NA12878). a Library yield was determined by Qubit and was directly correlated with DNA input amount for inputs smaller than 100 ng; the beads became saturated at 100 ng leading to normalized yields of around 11 ng/μl. Sequencing of the libraries on a MiSeq system and data analysis by the BaseSpace Whole Genome Sequencing app revealed consistent index representation across different DNA input amounts for multiple users (b) and a median insert length that was slightly smaller at lower DNA input amounts but was consistently around 350 bp for sample inputs of 100 ng and above (c). Libraries prepared from integrated DNA extraction protocols for blood (d) and saliva (e). f Library traces for low DNA input amounts. g There was almost complete overlap in single nucleotide variant (SNV) calls for libraries prepared from 0.5 ng and 100 ng inputs; each library was sequenced on one lane of a HiSeqX with data analysis performed using the BaseSpace Whole Genome Sequencing app and Variant Calling Assessment Tool

Fig. 2

Customization of post-amplification cleanup conditions. Libraries were prepared from 100 ng of human gDNA (NA12878), sequenced on a HiSeq X, and data analysis performed using the BaseSpace Whole Genome Sequencing 6.0.0 app. The median library fragment size and yield were modified by varying the volume of the Sample Preparation Beads (SPB) used during the first or second step of post-amplification cleanup. As extreme conditions were not sequenced, the median insert size was not available for all combinations. ND, not determined

Fig. 3

Improved coverage of human and bacterial genomes with Nextera DNA Flex. a Coverage across important regions of the human genome by three library preparation kits: Nextera DNA Flex, TruSeq Nano, and TruSeq PCR-free. b Coverage across extreme regions of the human genome by three library preparation kits: Nextera DNA Flex, TruSeq Nano, and TruSeq PCR-free. c Libraries generated from the small genomes of bacteria with low (B. cereus), medium (E. coli), and high (R. sphaeroides) GC content at 1 ng inputs. Less coverage variation was observed with libraries prepared by Nextera DNA Flex compared with libraries prepared by other commercially available library preparation kits (NEB, NEBNext Ultra DNA Library Prep Kit; Kapa A, Kapa HyperPlus Kits; Kapa B, Kapa HyperPrep Kits), particularly for the low and high GC content genomes of B. cereus and R. sphaeroides. The method of DNA fragmentation and whether PCR amplification was used during library preparation is indicated

Fig. 4

Application of Nextera DNA Flex to human amplicons. a Human leukocyte antigen (HLA) gene amplicons used as inputs for library preparation visualized on a 1% agarose gel. Lanes and expected amplicon sizes are as follows: 1, KBL Ladder; 2, HLA-A (4.1 kb); 3, HLA-B (2.8 kb); 4, HLA-C (4.2 kb); 5, HLA-DPA1 (10.3 kb); 6, HLA-DPB1 (9.7 kb); 7, HLA-DQA1 (7.3 kb); 8, HLA-DRB2 (4.6 kb); 9, HLA-DQB1 (7.1 kb). b Nextera DNA Flex library yields of all HLA amplicons were within the acceptable values of > 4 ng/μl and 9–13 ng/μl for 1 ng and 100–300 ng inputs, respectively. The yields for Nextera DNA Flex libraries were higher than for those prepared using TruSight HLA; for TruSight HLA, libraries were prepared from 1 ng of each amplicon and then pooled. c The Bioanalyzer profiles depict library fragment size distributions within the acceptable range; the distribution is narrower for the Nextera DNA Flex libraries (1 ng DNA inputs) than the TruSight HLA libraries. d Sequencing coverage depth and uniformity were higher for libraries prepared using Nextera DNA Flex (Flex) compared with TruSight HLA (TS HLA). e Libraries were sequenced on a NextSeq 550, with downsampling to 25,000 reads per amplicon. Library preparation using Nextera DNA Flex (orange) resulted in more uniform coverage of the entire human mitochondrial chromosome when compared with Nextera XT (grey). The location of the PCR primers used to create the two mtDNA amplicons are depicted by blue and red arrows. Dotted-line rectangle indicates the D-Loop region. f Zoomed in view shows more uniform coverage with Nextera DNA Flex within the D-Loop region

Fig. 5

Application of Nextera DNA Flex to bacterial amplicons. a Libraries prepared using Nextera DNA Flex showed more consistent, even coverage compared with libraries prepared using Nextera XT; data depicts the sequence coverage of libraries prepared from the 3 kb E. coli amplicon. b PCR products ranging in size from 50 bp to 3 kb amplified from E. coli gDNA visualized on a 1% agarose gel. c Libraries prepared from a 1 ng input of these E. coli amplicons resulted in Bioanalyzer traces that depicted a slight increase in fragment size with increasing amplicon size. d Libraries were sequenced on a MiSeq and coverage of the E. coli genome determined for the different amplicon fragment size inputs. Sequenceable libraries were generated from amplicons ranging in size from 50 bp to 3 kb. e When sequencing data was downsampled to 25,000 reads, the larger fragment inputs were reaching a coverage maximum

Fig. 6

Bioanalyzer traces of libraries prepared from various sample types and species. a Libraries prepared from samples with a varied degree of formalin fixation; a higher ΔCq indicates more FFPE-induced DNA degradation compared with a positive control. b Increasing FFPE-induced DNA degradation has a small effect on average fragment size but a marked effect on the total library yield. Increasing the DNA input from 100 ng to 150 ng did not increase library yield, indicating bead saturation at a DNA input of around 100 ng regardless of the degree of DNA degradation. c Libraries prepared from gDNA from a range of animal (human, Angus, and mouse), plant (Arabidopsis and alfalfa), and bacterial (E. coli and B. cereus) species

Fig. 7

Nextera DNA Flex Library Prep workflow overview. * Time estimates based on preparing 16 samples using a multichannel pipette. BLB, blood lysis buffer. BLT, bead-linked transposome. EPM, enhanced PCR mix. EtOH, Ethanol. PK1, proteinase K. RSB, resuspension buffer. SPB, sample purification beads. TB1, tagmentation buffer 1. TSB, tagment stop buffer. TWB, tagment wash buffer