Table 1 The feature set description
From: PESM: predicting the essentiality of miRNAs based on gradient boosting machines and sequences
Category | Description | Number of features |
|---|---|---|
Base content in pre-miRNAs | The content of base S in pre-miRNAs, S∈{U,C,G} | 3 |
mature-miRNAs length | The sequence length of mature-miRNAs | 1 |
Base content in mature-miRNAs | The content of base S in mature-miRNAs, S∈{U,C,G} | 3 |
non-mature-miRNAs length | The sequence length of non-mature-miRNAs | 1 |
Base content in non-mature-miRNAs | The content of base S in non-mature-miRNAs, S∈{U,C,G} | 3 |
MFE and nMFE | The minimum free energy of pre-miRNA secondary structures and it is divided by its length | 2 |
Cleavage site base class | The cleavage sites are assigned into 3 classes, 1: all cleavage sites of mature-miRNAs from the same pre-miRNAs are U; 0: not all cleavage sites are U; -1: all are non-U. | 1 |
Dinucleotide pairs frequency in pre-miRNAs | The Dinucleotide pairs SZ frequency in pre-miRNAs, S,Z∈{U,C,G} | 9 |
Dinucleotide pairs frequency in mature-miRNAs | The Dinucleotide pairs SZ frequency in mature-miRNAs, S,Z∈{U,C,G} | 9 |
The structure feature of pre-miRNAs | Normalized base-pairing propensity (P(s)), Normalized base-pairing propensity divided by its length (nP(s)), Normalized Shannon entropy (Q(s)), Normalized Shannon entropy divided by its length (nQ(s)), Normalized base-pair distance (D(s)), Normalized base-pair distance divided by its length (nD(s)) | 6 |