Algorithm 1. An enhanced positive-unlabeled learning algorithm.
Input	P—Positive training set; U—Unlabeled training set; $\partial$—The distance coefficient; $V_{s_i}$—Sequence $s_i$ in P and U;
	Model1,2,3,4,5—Five models trained by five subsets with P respectively;
	N1,2,3,4,5—Five negative sets predicted by Model1,2,3,4,5 on the remaining unlabeled training set respectively; cs—Common sequences of five negative sets N1,2,3,4,5 $N_{sv}$—Negative support vectors of five Model1,2,3,4,5
Output	F—Final classifier.
Stage 0:	Initialization
	l$\leftarrow$0; Avg_dist = 0; LN = ∅; RN = ∅; i
Stage 1:	Select the reliably negative initial set
	pr = ${\displaystyle \sum }_{\mathrm{i}}^{\left|P\right|}{V}_{s_i}/|P|$;
	Avg_dist + = ${{\displaystyle \sum }}_i^{\left|U\right|}dist(pr,V_{s_i})/|U|$;
	FOR i from 1 to |U|
	IF dist(pr,$V_{s_i}$) > Avg_dist * $\partial$
	LN = LN∪{Si};
	END IF
	END FOR
	Randomly divide the LN into five subsets D1, D2, D3, D4, D5.
	FOR i from 1 to 5
	Modeli = SVM(P, Di); Ni = Modeli(U − LN);
	END FOR
	The common sequence are represented to reliably negative initial set cs = N1 ∩ N2 ∩ N3 ∩ N4 ∩ N5; RN0 = RN0 ∪ cs; then the negative support vectors $N_{sv}$ of five models are included in ${RN}^0$ = ${RN}^0{{\displaystyle \cup }}^{}{N}_{sv}$.
Stage 2	Expand the reliably negative set
	WHILE TRUE
	IF Ul > 5∗|P|
	$U^{l+1}$ = $U^l$−$N_{pred}^l$;
	${RN}^{l+1}$ = $N_{pred}^l{{\displaystyle \cup }}^{}{N}_{sv}^l$;
	ELSE IF Ul < 5 ∗ |P|
	Go to Stage 3
	END IF
	Train a SVM classifier fl+1 on the P$\cup$RNl+1 with optimal parameter C and γ.
	Each sequence xi in Ul+1 would have a decision value f(xi) through the obtained fl+1, use the threshold T to get the reliably negative set.
	l $\leftarrow$ l + 1
Stage 3	Return the final classifier
	Return F = (P, RN)