Radiated Hybrid data sets

CARTHAGENE can handle both haploid and diploid radiated hybrid panels data using the so-called ``equal retention model'' [LBLC95]. CARTHAGENE uses a specific boosted EM [SCBM01] algorithm for handling haploid data. This algorithm may run one or two orders of magnitude faster than a standard EM algorithm without any loss of precision. There is also an experimental ``haploid model with errors'' which explicitely models errors in the data and which provides a posteriori probability for errors at the cost of speed (it is of the same order of efficiency as the diploid model). We therefore strongly recommend you to use the haploid model for all RH data and to shift to the diploid model only for final estimation of distances.

Depending on the model you intend to use (haploid or diploid), the first header line of the dataset should be respectively set to:

data type radiated hybrid

or

data type radiated hybrid error

or

data type radiated hybrid diploid

The character used to encode RH data are repectively:

• A for missing (Absent) marker.
• H for present (Here) marker.
• - for unknowns (eg. inconsistent typing when double typing is performed)