AbstractHypervariable, dominant molecular markers have become more common as characters for reconstructing evolutionary relationships, largely due to the difficulty in locating useful variation in DNA sequences in some very recently radiated taxa. Several of the methodological and analytical issues relating to these types of data are discussed, specifically focusing on inter‐simple sequence repeat (ISSR) data. Caution in avoiding redundancy in primer selection is recommended, as well as the use of 3' anchored primers. Manual versus automated methods are briefly discussed, in addition to the influence of fluorescent markers on ISSR PCR reactions. A series of clustering and phylogenetic analysis methods are compared. Several similarity coefficients are available for clustering analyses of ISSR data. Those that exclude shared‐absence characters appear most appropriate, such as Nei & Li's (1979) coefficient. However, more than one coefficient with this title is accessible from different software programs; care should be taken to be certain of the characteristics of a given coefficient. The potential suitability of Dollo, Fitch, or Camin‐Sokal Parsimony for ISSR analyses is also discussed.
Read full abstract