Loading and Merging data sets

To load a new dataset in CARTHAGENE, the only command is the dsload command which must be followed by the name of the file to load. The dataset loaded will get a numerical id (reported when the dataset is loaded and which can be requested later using the dsinfo command).

To do multiple populations mapping, two merging commands can be used. Each of these commands takes as arguments the numerical id of the 2 datasets merged (if you want to merge more than 2 datasets together, simply use the commands repeatedly).

• mergen: assumes that the markers order and the distance/recombination ratios are the same in the two datasets merged. This command cannot be used to merge radiated hybrid data with genetic data (for obvious reasons) and also cannot merge RIL self/sib data with other genetic data (for mathematical reasons). On such datasets, a single consensus map is reported by all mapping commands.
• mergor: assumes only that the markers order is the same in the two populations merged. Population specific distances will be estimated. Any pair of types of dataset can be merged by this command. On such dataset, two maps are reported by all mapping commands. When few markers are common to the merget datasets, many equivalent maps may populate the heap and limit the efficiency of the buildfw command. See the heapequiset command.

All mapping/map validating commands applied to merged datasets still compute exact multipoint maximum likelihood.

These two commands can be used together and repeatedly. Imagine for example that you have 2 radiated hybrid panels with the same irradiation level along with 2 genetic datasets available. You can ``mergen'' the 2 RH datasets together (assuming that the same irradiation level will lead to the same distances) and also ``mergen'' the 2 genetic datasets. This will give you two merged datasets which can be again merged using the mergor command. This gives a single dataset that merges all 4 datasets. All mapping commands on this dataset will report two maps: one RH consensus map and one genetic consensus map with the guarantee that they use a consistent markers order.