Table 3 |
|||||
|
Calculation of true and false positive and false negative rates from search results. |
|||||
|
Score cut-off1 |
BESs with positions |
BESs in tail-to-tail BACs |
BESs predicted to be in tail-to-tail BACs2 |
fp rate3 |
fn rate4 |
|
|
|||||
|
8 |
292,916 |
130,358 |
230,616 |
0.43 |
0.21 |
|
40 |
283,315 |
130,236 |
215,746 |
0.40 |
0.24 |
|
45 |
263,060 |
129,806 |
186,000 |
0.30 |
0.30 |
|
50 |
245,983 |
128,826 |
162,636 |
0.21 |
0.34 |
|
55 |
236,647 |
126,698 |
150,524 |
0.16 |
0.36 |
|
60 |
231,047 |
125,670 |
143,484 |
0.12 |
0.38 |
|
65 |
225,201 |
123,140 |
136,314 |
0.10 |
0.40 |
|
70 |
221,398 |
121,134 |
131,750 |
0.08 |
0.40 |
|
100 |
200,409 |
106,368 |
107,954 |
0.01 |
0.46 |
|
|
|||||
|
1results from the MegaBLAST search on full length BESs using a contiguous word size of 8 the default score cut-off is 8, the results in this row correspond to the case where no score cut-off is specified. 2for each score cut-off the number of BESs predicted to be in tail-to-tail BACs assuming all BESs were correctly positioned. 3false positive rate 4false negative rate |
|||||
|
Ratnakumar et al. BMC Genomics 2010 11:458 doi:10.1186/1471-2164-11-458 |
|||||