Genetic mapping of horizontal stripes in Lake Victoria cichlid fishes: benefits and pitfalls of using RAD markers for dense linkage mapping. Issue 21 (19th August 2014)

• Record Type: Journal Article
• Authors: Henning, Frederico; Lee, Hyuk Je; Franchini, Paolo; Meyer, Axel
• Is Part Of: Molecular ecology. Volume 23:Issue 21(2014)
• Page Start: 5224
• Page End: 5240
• Publication Date: 2014-08-19
• Subjects: Molecular ecology -- Periodicals; Molecular population biology -- Periodicals; 576
• DOI: 10.1111/mec.12860
• Languages: English
• ISSNs: 0962-1083

Abstract

<abstract abstract-type="main" id="mec12860-abs-0001"> <title>Abstract</title> <p>The genetic dissection of naturally occurring phenotypes sheds light on many fundamental and longstanding questions in speciation and adaptation and is a central research topic in evolutionary biology. Until recently, forward‐genetic approaches were virtually impossible to apply to nonmodel organisms, but the development of next‐generation sequencing techniques eases this difficulty. Here, we use the ddRAD‐seq method to map a colour trait with a known adaptive function in cichlid fishes, well‐known textbook examples for rapid rates of speciation and astonishing phenotypic diversification. A suite of phenotypic key innovations is related to speciation and adaptation in cichlids, among which body coloration features prominently. The focal trait of this study, horizontal stripes, evolved in parallel in several cichlid radiations and is associated with piscivorous foraging behaviour. We conducted interspecific crosses between <italic>Haplochromis sauvagei</italic> and <italic>H. nyererei</italic> and constructed a linkage map with 867 SNP markers distributed on 22 linkage groups and total size of 1130.63 cM. Lateral stripes are inherited as a Mendelian trait and map to a single genomic interval that harbours a paralog of a gene with known function in stripe patterning. Dorsolateral and mid‐lateral stripes were always coinherited and are thus under the same genetic control. Additionally, we directly quantify the genotyping error rates in RAD markers and offer guidelines for identifying and dealing with errors. Uncritical marker selection was found to severely impact linkage map construction. Fortunately, by applying appropriate quality control steps, a genotyping accuracy of >99.9% can be reached, thus allowing for efficient linkage mapping of evolutionarily relevant traits.</p> </abstract>