Abstract
The reference sequences of structurally complex regions can be obtained only through highly accurate clone-based approaches. We and others have successfully used single-haplotype iterative mapping and sequencing (SHIMS) 1.0 to assemble structurally complex regions across the sex chromosomes of several vertebrate species and to allow for targeted improvements to the reference sequences of human autosomes. However, SHIMS 1.0 is expensive and time consuming, requiring resources that only a genome center can provide. Here we introduce SHIMS 2.0, an improved SHIMS protocol that allows even a small laboratory to generate high-quality reference sequence from complex genomic regions. Using a streamlined and parallelized library-preparation protocol, and taking advantage of inexpensive high-throughput short-read-sequencing technologies, a small laboratory with both molecular biology and bioinformatics experience can sequence and assemble 192 large-insert bacterial artificial chromosome (BAC) or fosmid clones in 1 week. In SHIMS 2.0, in contrast to other pooling strategies, each clone is sequenced with a unique barcode, thus enabling clones containing nearly identical sequences to be multiplexed in a single sequencing run and assembled separately. Relative to SHIMS 1.0, SHIMS 2.0 decreases the required cost and time by two orders of magnitude while preserving high sequencing accuracy.