Filtered Targets Wizard

Open the Filtered Targets Wizard by going to Tools -> Manage Reference Samples... and then select the Create Low Quality Targets... button in the lower right hand corner. The filtered targets wizard uses aggregate statistics from precomputed CNV reference samples to create an annotation track with poor quality regions. This track can then be used to remove these regions from analysis when running the CNV algorithm in order to improve the accuracy and sensitivity of the calls.

Add Reference Set Files

The Sample Panel drop down looks in the default reference sample path by default. Each record set found in the folder is displayed in the drop down with the number of sample in that reference set. The reference sets are grouped based on the panel hash which is comprised of the regions in the panel as well as the options that were selected when coverage statistics was computed.

Set Target Region Filters

Multiple filters can be selected to determine the low quality targets in the reference panel. These are:

  • Low GC Content - This option will remove regions with a GC percentage below the specified threshold.
  • High GC Content - This option will remove the regions with a GC percentage above the specified threshold.
  • Short Span Regions - This will remove small regions.
  • Low Coverage - This will remove regions where the mean sample coverage is below the specified target threshold.
  • Overlapping Annotations - Adding annotation tracks allows for removal of targets that overlap regions where CNV detection is difficult. For example, low mapability regions could be removed since alignment in these regions is difficult leading to inconsistent coverage profiles.

By default the regions that pass the specified targets are removed from the track. This option can be disabled to include all of the regions in the reference panel. This can be helpful when fine tunning the filter thresholds

Confirm Target Track Options

This page show gives a summary of the filter options selected and allows you to set the:

  • Track Name - This is the internal track name, which will appear when plotting the source and when selecting the source as an annotation in VarSeq.
  • File Name - This is the name that will be used when writing the file.
  • Path This is the path where the new reference file will be written. If you would like to add the path to the Data Source Library for easy future reference check “Add Path to Library”.