Skip to main content
Figure 10 | Genome Biology

Figure 10

From: An optimization framework for unsupervised identification of rare copy number variation from SNP array data

Figure 10

Illustration of moves in the proposed iterative-improvement based optimization algorithm. The CNV assignment for a hypothetical region after t iterations is shown in the initial figure. Suppose Q ij is selected for the (t+1)st iteration. (a) All the markers in Q ij are assigned to C0, which completely eliminates Q ij as a gain region. This is a valid move. (b) Some markers that initially have a C0 class on the right border of Q ij are assigned to the C+ class which merges them with Q ij by a valid move. (c) Some markers that initially exist in Q ij are assigned to the C0 class, which contracts Q ij . This also represents a valid move. (d) An invalid move that divides the Q ij into two sub-gain regions. (e) Another invalid move that introduces a completely new gain region that is not identified in the previous solution.

Back to article page