Table 2.
Full-mutation samples detected in the EGA and simulated genomes using the default implementation of STR analysis tools
| Gene | AR (n = 1) | ATN1 (n = 2) | ATXN1 (n = 3) | ATXN3 (n = 1) | C9orf72 (n = 3) | DMPK (n = 17) | FMR1 (n = 18) | FMR2 (n = 3) | FXN (n = 25) | HTT (n = 13) | FPs | Total FM detected | Sensitivity |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| FM threshold (rpts) | 37 | 47 | 38 | 59 | 60 | 50 | 200 | 200 | 65 | 39 | |||
| Allelic classification | FM | FM | FM | FM | FM | FM | FM | FM | NL/FM or FM/FM | NL/FM or FM/FM | |||
| Isaac | |||||||||||||
| EH_v2 | 1 | 2 | 2 | 1 | 3 | 17 | 1 | 0 | 25 | 13 | 6 | 65 | 0.75581395 |
| EH_v3 | 1 | 2 | 3 | 0 | 3 | 17 | 0 | 0 | 25 | 13 | 5 | 64 | 0.74418605 |
| GangSTR | 0 | 2 | 2 | 0 | 0 | 16 | 0 | 0 | 16 | 11 | 8 | 47 | 0.54651163 |
| TRED | 1 | 2 | 1 | 0 | 3 | 17 | 0 | 0 | 25 | 13 | 3 | 62 | 0.72093023 |
| STRetch | 1 | 2 | 3 | 1 | 3 | 17 | 2 | 3 | 20 | 13 | 26 | 65 | 0.75581395 |
| exSTRa | 1 | 2 | 3 | 0 | 3 | 17 | 1 | 3 | 5 | 13 | 33 | 48 | 0.55813953 |
| BWA | |||||||||||||
| EH_v2 | 1 | 2 | 2 | 1 | 3 | 17 | 0 | 0 | 25 | 13 | 6 | 64 | 0.74418605 |
| EH_v3 | 1 | 2 | 3 | 0 | 3 | 17 | 0 | 0 | 25 | 13 | 5 | 64 | 0.74418605 |
| GangSTR | 1 | 2 | 2 | 1 | 1 | 16 | 0 | 0 | 0 | 10 | 8 | 33 | 0.38372093 |
| TRED | 1 | 2 | 1 | 0 | 3 | 17 | 0 | 0 | 25 | 13 | 10 | 62 | 0.72093023 |
| STRetch | 1 | 2 | 3 | 1 | 3 | 17 | 2 | 3 | 20 | 13 | 26 | 65 | 0.75581395 |
| exSTRa | 1 | 2 | 3 | 1 | 3 | 16 | 9 | 3 | 25 | 13 | 35 | 76 | 0.88372093 |
The analyzed dataset had 86 samples with at least one known full-mutation allele. The number of true-positives detected by the tools, sensitivity, and the number of false positives identified in our default analysis of the Isaac- (top panel) and BWA-aligned (bottom panel) genomes are shown. NL normal, FM full-mutation, FPs false-positives, rpts repeats, EH_v2 ExpansionHunter version 2, EH_v3 ExpansionHunter version 3, TRED TREDPARSE