Table 5. “Newly discovered” CNVs detected by each algorithm based on 48 normal and 48 schizophrenia subjects, using the “any” overlap criterion.
| Algorithm | CNVs Not Previously Discovered in Normals | CNVs Found Only in SCZ, Not Previously Discovered in Normals | CNVs Found Only in SCZ, Not Previously Discovered in Normals or SCZ | |||||||||
| Loss | Gain | Both | Total | Loss | Gain | Both | Total | Loss | Gain | Both | Total | |
| PennCNV | 72 | 84 | 33 | 189 | 24 | 31 | 2 | 57 | 24 | 29 | 2 | 55 |
| HMMSeg | 4 | 6 | 10 | 2 | 2 | 2 | 2 | |||||
| cnvPartition with 3 Probes | 12 | 2 | 14 | 3 | 3 | 3 | 3 | |||||
| cnvPartition with 5 Probes | 10 | 1 | 11 | 4 | 4 | 4 | 4 | |||||
| cnvPartition with 10 Probes | 2 | 2 | 1 | 1 | 1 | 1 | ||||||
| QuantiSNP | 1 | 1 | 0 | 0 | ||||||||
| HMMSeg & cnvPartition with 3 Probes | 1 | 1 | 0 | 0 | ||||||||
| HMMSeg & cnvPartition with 5 Probes | 1 | 1 | 0 | 0 | ||||||||
| HMMSeg & cnvPartition with 10 Probes | 0 | 0 | 0 | |||||||||
| HMMSeg & cnvPartition with 3 Probes & PennCNV & QuantiSNP | 1 | 0 | 0 | |||||||||
See Table 2 for explanation of algorithms.