Fast and robust imputation for miRNA expression data using constrained least squares

BMC Bioinformatics

Table 2 Singapore data results

Metric	FLI	VIPER	scImpute	Regression	Mean	Zeros
(a) Classification results
AUC	.95	.92	.91	.82	.80	.76
\(F_1\)	.91	.88	.88	.79	.80	.78
ACC	.89	.86	.87	.76	.74	.70
(b) Imputation errors
\(\epsilon _{\mu }\)	.35	.41	.45	.73	.75	.70
\(\epsilon _{\sigma }\)	.25	.34	.39	.54	.43	.06
\(\epsilon _{M}\)	1.12	1.94	2.50	3.53	1.85	.90

Time	FLI	VIPER	scImpute	Regression	Mean	Zeros
(c) Imputation time
\(t_\mu\)	.009	7.29	.100	\(\sim 0\)	\(\sim 0\)	\(\sim 0\)
\(t_\sigma\)	.002	8.42	.060	\(\sim 0\)	\(\sim 0\)	\(\sim 0\)
\(t_{\text {max}}\)	.027	49.2	.300	.003	.001	\(\sim 0\)

(a) Mean values over curves shown in Fig. 4a–c. (b) Mean values over curves shown in Fig. 4d–f. (c) Mean (\(t_\mu\)), standard deviation (\(t_\sigma\)), and maximum (\(t_{\text {max}}\)) imputation times (in seconds) over all test patients. In table (c), \(\sim\)0 indicates that the imputation time is strictly less than .0005 s

ISSN: 1471-2105