Back

A comparison between low-cost library preparation kits for low coverage sequencing

Stewart, C. M.; Gibson, M. J.; Parsa, J.-Y.; Li, J. H.

2024-01-31 genomics
10.1101/2024.01.30.578044 bioRxiv
Show abstract

In the fields of human health and agricultural research, low coverage whole-genome sequencing followed by imputation to a large haplotype reference panel has emerged as a cost-effective alternative to genotyping arrays for assaying large numbers of samples. However, a systematic comparison of library preparation methods tailored for low coverage sequencing remains absent in the existing literature. In this study, we evaluated one full sized kit from IDT and miniaturized and evaluated three Illumina-compatible library preparation kits--the KAPA HyperPlus kit (Roche), the DNA Prep kit (Illumina), and an IDT kit--using 96 human DNA samples. Metrics evaluated included imputation concordance with high-depth genotypes, coverage, duplication rates, time for library preparation, and additional optimization requirements. Despite slightly elevated duplication rates in IDT kits, we find that all four kits perform well in terms of imputation accuracy, with IDT kits being only marginally less performant than Illumina and Roche kits. Laboratory handling of the kits was similar: thus, the choice of a kit will largely depend on (1) existing or planned infrastructure, such as liquid handling capabilities, (2) whether a specific characteristic is desired, such as the use of full-length adapters, shorter processing times, or (3) use case, for instance, long vs short read sequencing. Our findings offer a comprehensive resource for both commercial and research workflows of low-cost library preparation methods suitable for high-throughput low coverage whole genome sequencing.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.