Figure 1.
(A) obtain a BAC library for the target organism; (B) select gene-enriched BACs from the library (optional); (C) fingerprint BACs and build a physical map; (D) select a minimum tiling path (MTP) from the physical map; (E) pool the MTP BACs according to the shifted transversal design; (F) sequence the DNA in each pool, trim/clean sequenced reads; (G) assign reads to BACs (deconvolution); (H) assemble reads BAC-by-BAC using a short-read assembler.
Figure 2.
An illustration of the three cases we are dealing with during the deconvolution process (clones belong to a MTP).
Figure 3.
Count distribution for the signatures of all distinct 26-mers [(a) rice synthetic data, (c) barley HV5] and all the reads [(b) rice synthetic data, (d) barley HV5] in the 91 pools of sequencing data.
The x-axis represents the size of the signature and the y-axis is the absolute count.
Table 1.
Summary of the statistics of the various assemblies obtained using Velvet (rows 1–3, 5–9) and SOAPdenovo (rows 4, 10, 11).