Predicting the Types of J-Proteins Using Clustered Amino Acids

Table 5

Results obtained in identifying functional types with (a) single amino acid case (𝑛=1) and (b) dipeptide case (𝑛=2).

(a) For the single amino acid case (𝑛=1)

Subfamily	Metrics	Feature dimension of 𝑛=1 for each cluster profile
		CP(20)	CP(13)	CP(11)	CP(9)	CP(8)	CP(5)
		20	13	11	9	8	5
Type I J-protein	Sn	71.42%	65.08%	68.25%	52.38%	50.79%	22.22%
	Sp	98.58%	98.66%	98.66%	98.93%	98.48%	98.03%
	MCC	0.71	0.67	0.69	0.60	0.56	0.26
Type II J-protein	Sn	33.96%	30.19%	33.96%	16.98%	16.98%	15.09%
	Sp	99.82%	99.21%	99.12%	99.47%	99.56%	99.09%
	MCC	0.54	0.42	0.45	0.30	0.31	0.23
Type III J-protein	Sn	98.74%	98.28%	98.10%	99.09%	98.73%	98.19%
	Sp	48.12%	42.86%	45.52%	32.31%	31.54%	17.46%
	MCC	59.71%	0.53	0.54	0.48	0.45	0.26
Type IV J-protein	Sn	4.54%	0	0	0	0	0
	Sp	99.91%	100.00%	100.00%	100.00%	100.00%	100.00%
	MCC	0.15	0.53	0	0	0	0
OA		92.93%	91.97%	92.13%	91.48%	91.08%	89.08%

(b) For the dipeptide case (𝑛=2)

Subfamily	Metrics	Feature dimension of 𝑛=2 for each cluster profile
		CP(20)	CP(13)	CP(11)	CP(9)	CP(8)	CP(5)
		400	169	121	81	64	25
Type I J-protein	Sn	74.42%	60.31%	73.02%	60.32%	58.73%	49.20%
	Sp	97.58%	98.59%	98.76%	97.71%	98.32%	97.79%
	MCC	0.75	0.63	0.73	0.58	0.60	0.5
Type II J-protein	Sn	39.76%	45.23%	39.62%	39.62%	35.84%	28.30%
	Sp	94.31%	99.29%	99.48%	99.03%	98.60%	97.99%
	MCC	0.57	0.57	0.54	0.49	0.42	0.31
Type III J-protein	Sn	98.88%	98.10%	98.82%	97.74%	98.01%	97.31%
	Sp	46.37%	50.74%	51.14%	50.79%	48.80%	40.34%
	MCC	60.08%	0.59	0.62	0.57	0.56	0.46
Type IV J-protein	Sn	13.16%	27.27%	0	22.73%	25.00%	9.09%
	Sp	99.91%	99.91%	100.00%	100.00%	100.00%	99.91%
	MCC	0.13	0.48	0	0.47	0.47	0.24
OA		91.47%	92.93%	93.25%	91.97%	92.04%	91.16%