Skip to main content

Table 3 Description of top 10 features from one-hot, PSDSP and KNF encodings

From: NmSEER V2.0: a prediction tool for 2′-O-methylation sites based on random forest and multi-encoding combination

Rank

In one-hot

PSDSP

KNF

1

T at 0 position

Dinucleotide at − 1 and 0 position

Frequency of GA

2

A at −2 position

Dinucleotide at 0 and + 1 position

Frequency of TG

3

C at −3 position

Dinucleotide at −2 and − 1 position

Frequency of AG

4

C at −1 position

Dinucleotide at −3 and − 2 position

Frequency of CT

5

G at −3 position

Dinucleotide at −5 and − 4 position

Frequency of GG

6

G at −1 position

Dinucleotide at −4 and − 3 position

Frequency of AA

7

G at −6 position

Dinucleotide at −9 and − 8 position

Frequency of CC

8

G at −9 position

Dinucleotide at −8 and − 7 position

Frequency of TC

9

G at −8 position

Dinucleotide at −10 and − 9 position

Frequency of CA

10

G at −10 position

Dinucleotide at −6 and − 5 position

Frequency of GC