Cotton mapping article

I am going to try and post an article a day that I have read and reviewed briefly here. Most will have to do with my work.

This article looked at two maps: BC1F1 TM-1(G. hirsutum) x [TM-1 x Hai7124 (G. barbadense)], and Vsg (G. barbadense) x [TM-1 x Hai7124]. Vsg is a virescent, a pale yellow-green, semigamous barbadense line. Semigamy is a mutation where the sperm and the egg nucleus do not fuse at fertilization and so depending on orientation of the spindles can produce progeny that are haploid, diploid, or with sectors of both. Haploids may be maternal or paternal. Sectors may be haploid or diploid. Diploid sectors are most vigorous and may overcome haploid regions.

The authors claim the maps are highly collinear. Discrepancies are noted as minor statistical errors or difficulty in ordering markers due to high number of double cross overs. I disagree with this claim. The differences are significant. I am posting an example linkage group. This may be due to using semigamous line or for comparing BC1s with different recurrent species. Was DNA extracted from whole plants? All lines tested in semigamous cross were haploid/doubled. How many were truly doubled? How many paternal/maternal haploids were produced? Maternal haploids would have no linkage information. Only paternally derived haploids would be beneficial since F1 was paternal. Were they chimaeric?

Maybe we should all do like in the Quadmap paper and bootstrap each linkage group to get a sense of how reliable the associations are.

Xianliang Song, Kai Wang, Wangzhen Guo, Jun Zhang, and Tianzhen Zhang. 2005. A comparison of genetic maps constructed from haploid and BC1 mapping populations from the same crossing between Gossypium hirsutum L. and Gossypium barbadense L. Genome 48: 378–390