Table 2. 2×2 contingency table to apply χ2 test to identify significant sequence determinants.
| Count of a target sequence fragment1 |
Count of the other sequence fragments1 | Total count | |
|---|---|---|---|
| Foreground | N11 (N1+× N+1/N++) | N12 (N1+× N+2/N++) | N1+ |
| Background | N21 (N2+× N+1/N++) | N22 (N2+× N+2/N++) | N2+ |
| Total count | N+1 | N+2 | N++ |
1Numbers in the parenthesis are expected numbers to calculate the χ2 test statistics. For example, for a sequence fragment (e.g.”AACCGGTT”), N11 is its observed count in the foreground regions, N12 is the observed count of sequence fragments that are not “AACCGGTT”, but has the same length in the foreground regions, and N21 and N22 are the counterpart of the N11 and N12 in the background regions, respectively. The sign “+” means row-wise sums (N1+, N2+), column-wise sums (N+1, N+2), or total sum (N++). OR is estimated as N11×N22/(N12×N21).