Table 2 Process of stepwise character selection based on RF

Algorithm: Stepwise Character Selection
Input: Ranked independent gene list G, number of candidate genes n, gene expression microarray
DR(nd),threshold ε.
Output: Predictor set P, predict accuracy ACC,P_ACC.
Step1: Initialization: P=, candidate gene set C=G.
Step 1.1:Calculate the accuracy ACCi, SNi,SPi,MCCi by Eqs. 1, 2, 3, 4 of each ciC
which acts as a single predictor in RF with 5-fold cross validation, respectively.
Step 1.2: Pp,where $$p \leftarrow \mathop {argmax}\limits _{c_{i} \in C}ACC$$, $$P\_ACC \leftarrow \mathop {max}\limits _{i\in {1 \cdots n}}ACC,C \leftarrow C/p$$.
Step2: Character selection.
While P_ACCACC_max>−ε and C, do
Step2.1: ACC_maxP_ACC.
1.$$P_{add\_i} \leftarrow P \cup \{ c_{i} \in C\}$$ calculate $${ACC}_{add\_i}$$ using $$P_{add\_i}$$,as predictors in RF with 5-folf cross validation, i=1,,n.
2.$$P \leftarrow P_{add\_I_{add}}$$,where $$I_{add} \leftarrow \mathop {argmax}\limits _{i=1,\cdots,n}({ACC}_{add}),P\_ACC \leftarrow max(ACC),C \leftarrow C/P$$.
2.$$P_{remove\_i} \leftarrow P/p_{i} \in P$$,calculate $${ACC}_{remove\_i}$$ using $$P_{remove\_i}$$as predictors in Random Forest with 5-fold cross validation, i=1,,nremove.
3.$$P\leftarrow \{p_{i}|{ACC}_{remove\_i}>P\_ACC,i=1,\cdots,n_{remove}\}$$.
$$C \leftarrow \{C \cup \{p_{i}|{ACC}_{remove\_i} \leq P\_ACC,i=1,\cdots,n_{remove}\}$$ 