Table 2

The workflow of window size selection and parameter optimization.

Sliding window size (w)

C and γ

Smoothing window size (ws)

Weight parameter (w1 and w-1)


Step 1

3 ≤ w ≤ 41 (step = 2)

Default

-

Default ratio


Step 2

Optimized w from step 1

-3 ≤ log2C ≤ 12 (step = 1)

Default ratio

-3 ≤ log2γ ≤ -15 (step = -1)


Step 3

Optimized w from step 1

Optimized C and γ from step 2

3 ≤ ws ≤ 11 (step = 2)

Default ratio


Step 4

Optimized w from step 1

Optimized C and γ from step 2

Optimized ws from step 3

1 ≤ w1 ≤ 8# (step = 1), w-1 = 1


Final

Optimized w from step 1

Optimized C and γ from step 2

Optimized ws from step 3

Optimized w1 and w-1 from step4


# In the RBP107 data set, we test w1 ranged from 1 to 10 (step = 1).

Cheng et al. BMC Bioinformatics 2008 9(Suppl 12):S6   doi:10.1186/1471-2105-9-S12-S6

Open Data