Table 2.
PLS1 method to estimate the correct classification (%) of virus infection in E.festucae isolates
| Data pre-treatment | 2006 (n = 11) |
2008 (n = 31) |
2006 + 2008 (n = 42) |
||||||
|---|---|---|---|---|---|---|---|---|---|
| Fa | Virus (n = 6) |
No virus (n = 5) |
Fa | Virus (n = 16) |
No virus (n = 15) |
Fa | Virus (n = 22) |
No virus (n = 20) |
|
| Raw | 4 | 100 | 80 | 1 | 75 | 47 | 6 | 73 | 65 |
| MSCb | 3 | 100 | 80 | 4 | 81 | 60 | 5 | 77 | 70 |
| SNVc | 3 | 100 | 80 | 4 | 81 | 60 | 6 | 82 | 75 |
| 1Dd | 2 | 100 | 80 | 1 | 62 | 47 | 5 | 73 | 75 |
| MSCb+1Dd | 2 | 100 | 80 | 2 | 81 | 67 | 4 | 73 | 75 |
| SNVc+1Dd | 2 | 100 | 80 | 2 | 81 | 67 | 3 | 86 | 75 |
| 2De | 4 | 100 | 100 | 1 | 62 | 53 | 4 | 73 | 65 |
| MSCb+2De | 1 | 100 | 100 | 3 | 75 | 60 | 4 | 73 | 55 |
| SNVc+2De | 1 | 100 | 100 | 3 | 75 | 60 | 4 | 82 | 75 |
Results of the external validation for individual and combined years (2006 and 2008) using different data pre-treatments. Bold type indicates the best classifications obtained for detection of virus infection
a Number of PLS factors
b Multiplicative scatter correction
c Standard normal variate
d First derivative
e Second derivative