18.5.2 Including or excluding regions (masking)

The next part of the dialog lets you mask the reference sequences. Masking refers to a mechanism where parts of the reference sequence are not considered in the mapping. This can be extremely useful for example when mapping human data, where more than 50 % of the sequence consists of repeats. Note that you should be careful masking all the repeat regions if your sequenced data contains the repeats. If you do that, some of the reads that would have matched a masked repeat region perfectly may be placed wrongly at another position even with a less-perfect match and lead to wrong results.

In order to mask e.g. repeat regions when doing read mapping, the repeat regions have to be annotated on the reference sequences.

Because the masking is based on annotations, any kind of annotations can be selected for masking. This means that you can choose to e.g. only map against the genes in the genome, or only the exons. As long as the reference sequences contain the relevant information in the form of annotations, it can be masked.

To mask a reference sequence, first click the Include / exclude regions checkbox, and second click the Select annotation type (Image add_annotation_button) button.

This will bring up a dialog with all the annotation types of the reference sequences listed to the left. Select one or more annotation types and click Add (Image add) button. Then select at the bottom whether you wish to Include or Exclude these annotation types. If you include, it means that only the regions covered by the selected type of annotations will be used in the read mapping. If you exclude, it means that all of the reference sequences except the regions covered by the selected type of annotations will be used in the read mapping.

You can see an example in figure 18.50.

Image referenceassembly_mask
Figure 18.50: Masking for repeats. The repeat region annotation type is selected and excluded in the mapping.