xi wangab, erin gardinerabc, murray j. cairns abc · mol. biosyst., 2014 wang x et al....
TRANSCRIPT
Mol. BioSyst., 2014
Wang X et al.
Supplementary Materials for
Optimal consistency in microRNA expression analysis using reference-gene-based normalization
Xi Wangab, Erin Gardinerabc, Murray J. Cairns*abc
a School of Biomedical Sciences and Pharmacy, Faculty of Health and Medicine, The
University of Newcastle, NSW 2308, Australiab Hunter Medical Research Institute, Newcastle, NSW 2305, Australiac Schizophrenia Research Institute, Sydney, NSW 2010, Australia
* Corresponding author. Email: [email protected]
1
Electronic Supplementary Material (ESI) for Molecular BioSystems.This journal is © The Royal Society of Chemistry 2015
Mol. BioSyst., 2014
Wang X et al.
Supplementary Figure S1. Consistency score comparison between different normalization methods with SAM for DE analysis. (a) without ComBat; (b) with ComBat for batch effect adjustment.
2
(a)
(b)
Mol. BioSyst., 2014
Wang X et al.
(a) (b)
(c)
Supplementary Figure S2. Consistency scores based on array-matrix partitions. (a) F-test, (b) limma, and (c) SAM for DE analysis.
3
Mol. BioSyst., 2014
Wang X et al.
Supplementary Figure S3. Illustration of the calculation of consistency scores. The figures indicate the weights corresponding to each area. (a) is for three partitions, and (b) is for four partitions.
4
(a) (b)
Mol. BioSyst., 2014
Wang X et al.
Supplementary Table S1. The number of case and control samples distributed in each array matrix.
Array Matrix # Total Samples # Cases # ControlsNo. 1 87 45 42No. 2 59 48 11No. 3 42 23 19No. 4 23 16 7Total 211 128 83
5
Mol. BioSyst., 2014
Wang X et al.
Supplementary Table S2. Consistency scores for 12 combinations of transformation and normalization methods based on F-test DE analysis.
#(top genes) 5 10 20 50 100 150 200 250VST+QN 0.17(0.04) 0.76(0.08) 2.31(0.14) 13.95(0.34) 51.74(0.71) 111.28(0.97) 191.29(1.15) 289.62(1.36)VST+RSN 0.22(0.04) 0.86(0.08) 2.86(0.15) 15.15(0.38) 53.67(0.68) 112.42(0.97) 192.49(1.16) 288.95(1.30)VST+SSN 0.34(0.06) 1.08(0.10) 3.42(0.19) 15.52(0.40) 53.32(0.71) 113.21(1.11) 193.87(1.27) 292.08(1.42)VST+VSN 0.13(0.04) 0.69(0.08) 2.57(0.15) 14.07(0.38) 52.47(0.90) 110.11(1.34) 188.52(1.68) 284.13(1.87)VST+RIN 0.16(0.04) 0.57(0.07) 2.19(0.14) 12.15(0.38) 46.46(0.73) 103.68(1.13) 182.68(1.66) 283.34(2.34)VST+Loess 0.21(0.04) 0.77(0.09) 2.52(0.15) 14.59(0.36) 52.34(0.73) 112.53(0.98) 192.94(1.32) 289.70(1.44)Log+U49 0.37(0.05) 1.01(0.09) 3.21(0.16) 15.13(0.40) 52.99(0.72) 112.44(0.96) 191.22(1.16) 288.47(1.40)Log+U24 0.25(0.05) 0.92(0.09) 2.92(0.18) 15.10(0.43) 52.83(0.97) 111.06(1.52) 188.97(1.93) 285.57(2.15)Log+U49U24 0.37(0.06) 1.18(0.11) 3.45(0.17) 15.24(0.45) 52.25(0.85) 109.92(1.28) 187.72(1.63) 285.46(1.79)VST+U49 0.44(0.06) 1.59(0.13) 5.00(0.24) 22.94(0.75) 71.19(1.77) 136.43(2.63) 218.51(3.42) 313.63(4.10)VST+U24 0.54(0.07) 2.02(0.17) 7.41(0.35) 30.56(1.06) 87.73(2.50) 159.47(3.86) 244.48(5.31) 341.89(6.35)VST+U49U24 0.60(0.07) 2.29(0.17) 6.95(0.32) 28.30(0.97) 81.61(2.31) 150.70(3.49) 234.29(4.71) 331.48(5.71)
The figures outside parentheses are averaged scores from 100 data random partitions and the figures insides parentheses are standard errors (s.e.). The highest scores are in bold and the second highest scores in italic.
6
Mol. BioSyst., 2014
Wang X et al.
Supplementary Table S3. Consistency scores for 12 combinations of transformation and normalization methods based on limma DE analysis
#(top genes) 5 10 20 50 100 150 200 250VST+QN 0.26(0.05) 0.84(0.08) 2.55(0.15) 14.47(0.37) 53.19(0.73) 112.94(0.95) 193.38(1.16) 291.50(1.37)VST+RSN 0.30(0.05) 1.02(0.09) 3.14(0.18) 15.69(0.40) 55.07(0.72) 114.22(0.94) 194.60(1.19) 291.50(1.28)VST+SSN 0.35(0.06) 1.20(0.10) 3.63(0.19) 16.66(0.40) 55.05(0.74) 115.27(1.06) 196.00(1.29) 294.58(1.45)VST+VSN 0.24(0.05) 0.86(0.09) 2.91(0.16) 15.10(0.40) 53.33(0.92) 110.71(1.31) 189.45(1.61) 284.51(1.79)VST+RIN 0.20(0.04) 0.64(0.07) 2.40(0.16) 12.39(0.39) 46.83(0.73) 104.30(1.15) 183.33(1.70) 283.99(2.36)VST+Loess 0.28(0.05) 0.81(0.09) 2.74(0.15) 15.20(0.37) 53.40(0.76) 113.80(1.02) 194.53(1.34) 292.14(1.48)Log+U49 0.39(0.05) 1.27(0.10) 3.72(0.16) 16.08(0.41) 54.88(0.72) 114.60(0.99) 193.19(1.17) 290.15(1.41)Log+U24 0.33(0.06) 1.02(0.10) 3.24(0.18) 15.68(0.46) 53.61(1.00) 112.15(1.54) 189.70(1.94) 286.63(2.15)Log+U49U24 0.46(0.07) 1.33(0.12) 4.05(0.19) 16.41(0.50) 54.37(0.87) 112.09(1.33) 189.17(1.59) 286.99(1.76)VST+U49 0.54(0.07) 1.75(0.13) 5.44(0.26) 23.93(0.72) 73.13(1.76) 138.97(2.66) 221.55(3.47) 317.00(4.17)VST+U24 0.75(0.08) 2.91(0.20) 8.46(0.38) 34.12(1.09) 94.28(2.50) 168.95(3.96) 254.70(5.37) 353.62(6.49)VST+U49U24 0.65(0.08) 2.58(0.18) 7.57(0.33) 30.12(1.02) 84.78(2.32) 154.73(3.59) 238.03(4.77) 334.25(5.75)
The figures outside parentheses are averaged scores from 100 data random partitions and the figures insides parentheses are standard errors (s.e.). The highest scores are in bold and the second highest scores in italic.
7
Mol. BioSyst., 2014
Wang X et al.
Supplementary Table S4. Consistency scores for 12 combinations of transformation and normalization methods based on ComBat batch effect adjustment and F-test DE analysis.
#(top genes) 5 10 20 50 100 150 200 250VST+QN 0.12(0.03) 0.64(0.09) 2.55(0.17) 14.65(0.41) 54.68(0.78) 116.05(1.04) 196.14(1.21) 294.64(1.38)VST+RSN 0.13(0.04) 0.60(0.07) 2.83(0.16) 15.08(0.40) 54.71(0.82) 116.77(1.17) 197.84(1.36) 296.12(1.49)VST+SSN 0.27(0.06) 0.90(0.11) 3.42(0.20) 18.35(0.55) 60.70(1.14) 122.48(1.31) 203.80(1.57) 302.10(1.72)VST+VSN 0.14(0.03) 0.57(0.07) 2.04(0.14) 12.60(0.41) 49.98(0.86) 109.16(1.21) 188.02(1.53) 286.10(1.69)VST+RIN 0.30(0.06) 1.03(0.11) 4.01(0.23) 20.47(0.66) 65.65(1.39) 129.31(2.10) 209.28(2.78) 307.33(3.29)VST+Loess 0.23(0.04) 0.77(0.09) 2.47(0.17) 15.18(0.43) 55.11(0.78) 117.15(1.00) 197.46(1.23) 296.15(1.43)Log+U49 0.18(0.04) 0.78(0.10) 2.71(0.17) 15.52(0.50) 54.24(0.97) 115.47(1.28) 195.59(1.57) 294.95(1.78)Log+U24 0.33(0.05) 0.92(0.09) 3.12(0.19) 16.51(0.52) 59.03(1.11) 120.80(1.73) 201.36(2.23) 300.87(2.56)Log+U49U24 0.25(0.05) 0.79(0.09) 2.95(0.21) 15.84(0.55) 55.04(1.12) 116.21(1.52) 194.89(1.95) 291.85(2.12)VST+U49 0.22(0.05) 1.10(0.12) 3.61(0.21) 19.20(0.71) 65.02(1.80) 128.61(2.77) 208.99(3.51) 306.39(4.06)VST+U24 0.35(0.07) 1.27(0.11) 4.18(0.22) 20.63(0.72) 68.61(1.86) 135.61(2.94) 218.73(3.89) 315.62(4.45)VST+U49U24 0.24(0.05) 1.12(0.11) 3.82(0.23) 19.69(0.74) 66.68(1.87) 132.34(3.02) 213.84(3.99) 311.41(4.65)
The figures outside parentheses are averaged scores from 100 data random partitions and the figures insides parentheses are standard errors (s.e.). The highest scores are in bold and the second highest scores in italic.
8
Mol. BioSyst., 2014
Wang X et al.
Supplementary Table S5. Consistency scores for 12 combinations of transformation and normalization methods based on ComBat batch effect adjustment and limma DE analysis.
#(top genes) 5 10 20 50 100 150 200 250VST+QN 0.14(0.04) 0.72(0.09) 2.87(0.19) 15.53(0.46) 56.27(0.87) 118.56(1.11) 199.11(1.27) 297.26(1.46)VST+RSN 0.16(0.04) 0.70(0.09) 3.08(0.18) 16.01(0.43) 57.05(0.93) 118.83(1.21) 200.44(1.44) 298.73(1.57)VST+SSN 0.28(0.06) 0.97(0.11) 3.78(0.21) 19.25(0.61) 62.60(1.19) 124.89(1.36) 206.39(1.61) 304.64(1.84)VST+VSN 0.19(0.04) 0.62(0.07) 2.36(0.15) 12.98(0.37) 50.89(0.86) 109.89(1.26) 188.86(1.51) 285.88(1.72)VST+RIN 0.33(0.06) 1.29(0.12) 4.44(0.24) 22.11(0.69) 69.25(1.46) 133.24(2.18) 214.28(2.85) 311.50(3.36)VST+Loess 0.20(0.04) 0.87(0.10) 2.79(0.18) 16.09(0.46) 56.74(0.86) 119.11(1.02) 199.32(1.28) 298.39(1.39)Log+U49 0.20(0.04) 0.78(0.10) 2.78(0.18) 15.70(0.53) 54.68(1.02) 115.11(1.29) 195.78(1.63) 294.01(1.78)Log+U24 0.33(0.05) 0.97(0.09) 3.33(0.21) 17.45(0.57) 60.99(1.17) 124.11(1.80) 205.51(2.31) 305.62(2.77)Log+U49U24 0.28(0.05) 0.88(0.10) 3.17(0.22) 16.82(0.59) 57.02(1.22) 118.95(1.62) 198.06(2.05) 295.16(2.22)VST+U49 0.27(0.05) 1.11(0.12) 4.06(0.23) 20.37(0.75) 66.66(1.87) 131.03(2.79) 211.55(3.51) 308.34(4.00)VST+U24 0.52(0.08) 1.52(0.13) 4.96(0.28) 23.95(0.80) 75.66(1.94) 143.36(3.08) 227.28(3.94) 325.30(4.64)VST+U49U24 0.32(0.05) 1.14(0.11) 4.14(0.25) 21.24(0.78) 69.30(1.88) 135.55(3.01) 217.46(3.95) 314.98(4.53)
The figures outside parentheses are averaged scores from 100 data random partitions and the figures insides parentheses are standard errors (s.e.). The highest scores are in bold and the second highest scores in italic.
9
Mol. BioSyst., 2014
Wang X et al.
Supplementary Table S6. P-values of paired one-tailed t-tests for testing the mean difference of consistency scores between Method_1 and Method_2 based on F-test DE analysis. The alternative hypothesis was that the mean consistency score of Method_1 was higher than that of Method_2.
Method_1 Method_2 Top_5 10 20 50 100 150 200 250VST+QN 2.9E-05 1.8E-08 2.4E-17 2.5E-21 2.2E-16 4.7E-14 1.9E-11 3.2E-07VST+RSN 1.2E-03 7.2E-06 1.4E-10 5.1E-15 3.8E-14 1.0E-13 1.3E-10 1.2E-07VST+SSN 6.1E-02 4.7E-04 2.4E-07 5.2E-16 7.0E-15 5.2E-13 4.6E-10 1.5E-06VST+VSN 4.3E-06 1.6E-08 4.2E-13 1.3E-18 3.8E-15 2.3E-15 6.2E-14 2.3E-10VST+RIN 1.0E-04 1.3E-11 1.3E-17 4.7E-25 1.0E-23 1.1E-21 4.4E-17 7.3E-10VST+Loess 6.1E-04 4.3E-08 2.7E-14 9.9E-20 4.0E-16 2.8E-13 4.4E-10 2.9E-07Log+U49 2.2E-01 4.3E-04 3.5E-08 5.3E-16 3.0E-16 3.9E-13 1.2E-11 8.3E-08Log+U24 2.6E-03 2.7E-05 4.3E-10 3.2E-15 6.8E-15 4.0E-13 6.9E-12 1.1E-08Log+U49U24 1.4E-01 3.7E-03 2.5E-06 1.8E-15 5.5E-17 3.2E-14 9.8E-13 5.7E-09VST+U49 - - - - - - - -VST+U24 8.5E-01 9.6E-01 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00
VST+U49
VST+U49U24 9.7E-01 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00VST+QN 2.0E-06 2.8E-11 5.0E-30 2.2E-34 4.9E-28 4.4E-22 8.8E-17 1.2E-12VST+RSN 7.5E-05 2.4E-09 3.6E-25 1.3E-31 1.2E-26 5.4E-22 2.9E-16 4.0E-13VST+SSN 1.3E-02 1.8E-06 1.3E-19 1.9E-30 2.2E-28 1.2E-21 1.4E-15 8.1E-12VST+VSN 7.4E-08 1.2E-12 1.8E-27 1.1E-31 1.9E-27 7.7E-24 2.6E-18 1.6E-14VST+RIN 9.4E-07 6.0E-14 4.2E-30 3.5E-40 1.3E-35 4.9E-29 1.5E-20 2.6E-14VST+Loess 5.3E-05 1.2E-10 1.9E-29 2.0E-33 8.5E-28 8.6E-22 6.9E-16 2.4E-12Log+U49 2.6E-02 8.6E-07 4.7E-20 5.9E-31 3.2E-27 1.3E-21 7.6E-17 9.7E-13Log+U24 5.0E-05 4.2E-09 4.2E-25 1.9E-32 1.0E-30 2.7E-24 4.8E-19 5.9E-15Log+U49U24 1.0E-02 1.8E-06 8.8E-22 1.3E-32 4.6E-31 7.0E-24 8.5E-19 1.9E-14VST+U49 1.5E-01 3.9E-02 3.9E-11 2.1E-11 3.0E-11 4.4E-10 7.6E-08 1.8E-07VST+U24 - - - - - - - -
VST+U24
VST+U49U24 7.8E-01 9.9E-01 1.0E-02 1.5E-04 1.6E-07 2.4E-06 1.3E-05 5.4E-05VST+QN 6.0E-08 1.8E-16 3.1E-27 2.0E-30 4.5E-23 2.1E-19 1.9E-14 1.6E-10VST+RSN 9.0E-06 2.2E-13 2.3E-21 2.2E-26 2.4E-21 4.4E-19 8.1E-14 5.5E-11VST+SSN 2.1E-03 2.1E-09 1.9E-17 5.1E-26 5.2E-23 1.6E-18 3.4E-13 8.4E-10VST+VSN 5.9E-09 1.9E-15 1.8E-23 3.5E-27 1.4E-22 1.6E-20 3.8E-16 1.6E-12VST+RIN 4.7E-08 2.3E-17 2.3E-27 4.0E-35 3.5E-30 6.5E-26 1.2E-18 3.1E-12VST+Loess 1.8E-06 1.4E-14 4.2E-25 3.3E-29 7.6E-23 8.6E-19 2.6E-13 2.7E-10Log+U49 4.9E-03 4.8E-10 4.1E-17 4.2E-27 1.2E-22 9.3E-19 1.7E-14 1.1E-10Log+U24 3.0E-06 7.8E-13 3.9E-22 5.5E-27 1.0E-23 2.5E-19 1.1E-15 3.8E-12Log+U49U24 6.9E-04 1.8E-09 1.3E-18 9.1E-28 6.8E-26 3.1E-20 4.3E-16 3.8E-12VST+U49 3.0E-02 1.8E-04 8.3E-10 1.9E-10 3.6E-09 1.0E-08 2.7E-06 3.4E-06VST+U24 2.2E-01 1.3E-02 9.9E-01 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00
VST+U49U24
VST+U49U24 - - - - - - - -
10
Mol. BioSyst., 2014
Wang X et al.
Supplementary Table S7. P-values of paired one-tailed t-tests for testing the mean difference of consistency scores between Method_1 and Method_2 based on limma DE analysis. The alternative hypothesis was that the mean consistency score of Method_1 was higher than that of Method_2.
Method_1 Method_2 Top_5 10 20 50 100 150 200 250VST+QN 3.4E-04 1.1E-09 3.8E-16 6.8E-23 6.2E-17 8.4E-15 7.7E-12 4.8E-08VST+RSN 2.5E-03 6.6E-06 7.9E-10 1.2E-16 1.1E-14 1.6E-14 4.7E-11 5.8E-08VST+SSN 1.3E-02 6.0E-05 3.2E-09 2.7E-17 9.2E-16 7.1E-14 1.6E-10 5.3E-07VST+VSN 3.4E-04 2.0E-08 4.7E-12 5.9E-21 7.8E-17 5.5E-17 4.3E-15 7.7E-12VST+RIN 4.1E-05 2.2E-14 7.6E-16 1.5E-27 4.6E-26 2.1E-23 5.4E-18 4.5E-11VST+Loess 1.3E-03 6.4E-10 1.1E-13 3.1E-22 2.1E-17 3.1E-14 4.5E-11 1.1E-07Log+U49 6.5E-02 2.2E-03 2.1E-06 3.1E-17 7.9E-16 1.3E-13 4.8E-12 1.3E-08Log+U24 6.9E-03 2.3E-06 2.1E-09 4.4E-17 2.6E-16 4.2E-14 4.8E-13 8.3E-10Log+U49U24 1.8E-01 9.8E-04 9.9E-05 1.4E-15 3.6E-17 1.9E-14 1.6E-13 6.8E-10VST+U49 - - - - - - - -VST+U24 9.7E-01 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00
VST+U49
VST+U49U24 8.9E-01 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00VST+QN 1.4E-06 2.7E-20 9.7E-31 8.1E-38 1.8E-31 5.6E-26 8.3E-20 4.6E-16VST+RSN 6.3E-06 3.8E-19 2.8E-27 2.1E-34 2.8E-31 2.1E-26 2.0E-19 5.7E-16VST+SSN 1.3E-04 1.9E-14 2.1E-24 3.3E-33 6.8E-32 7.3E-26 1.2E-18 9.7E-15VST+VSN 3.9E-07 1.1E-20 2.3E-28 1.3E-38 2.8E-32 1.1E-27 1.3E-21 1.9E-18VST+RIN 1.4E-07 5.5E-23 1.4E-30 1.7E-43 2.8E-40 5.7E-33 4.5E-24 1.0E-17VST+Loess 4.8E-06 3.0E-21 8.8E-33 1.1E-37 4.4E-32 8.4E-26 3.0E-19 2.5E-15Log+U49 3.1E-04 3.7E-13 3.0E-22 7.3E-37 1.4E-30 1.1E-25 6.2E-20 4.1E-16Log+U24 3.7E-06 1.3E-20 3.1E-30 4.6E-37 4.1E-35 2.0E-28 6.9E-23 2.0E-18Log+U49U24 1.1E-03 9.0E-16 2.0E-23 2.9E-35 9.7E-35 1.0E-27 3.7E-22 1.1E-17VST+U49 2.7E-02 6.5E-09 6.4E-13 2.7E-18 1.9E-17 1.1E-15 5.8E-12 2.7E-11VST+U24 - - - - - - - -
VST+U24
VST+U49U24 1.2E-01 3.4E-03 6.1E-04 4.1E-09 3.6E-15 1.1E-13 2.5E-12 2.7E-13VST+QN 1.2E-05 2.7E-16 1.5E-31 3.5E-31 5.8E-24 7.7E-20 1.2E-14 9.2E-11VST+RSN 3.1E-04 3.0E-14 2.1E-25 5.7E-26 1.4E-22 9.7E-20 3.6E-14 1.0E-10VST+SSN 1.0E-03 2.7E-10 2.7E-22 3.0E-25 4.6E-24 2.3E-19 1.2E-13 9.2E-10VST+VSN 6.8E-06 4.4E-15 4.2E-24 1.0E-29 7.4E-25 1.3E-21 8.1E-17 1.9E-13VST+RIN 1.1E-06 9.7E-19 5.0E-29 5.6E-35 2.9E-32 1.0E-26 2.9E-19 9.1E-13VST+Loess 3.8E-05 9.4E-17 8.6E-31 8.8E-31 1.6E-24 2.3E-19 5.7E-14 3.3E-10Log+U49 4.0E-03 2.0E-09 1.5E-19 2.7E-28 3.6E-23 5.5E-19 9.7E-15 5.3E-11Log+U24 6.8E-05 1.2E-14 3.4E-26 1.9E-28 1.1E-25 2.0E-20 9.6E-17 1.4E-12Log+U49U24 1.1E-02 3.0E-11 5.0E-22 8.2E-28 1.5E-26 1.4E-20 1.1E-16 2.3E-12VST+U49 1.1E-01 1.1E-05 1.2E-11 2.8E-13 2.8E-11 4.0E-10 6.3E-07 5.9E-06VST+U24 8.8E-01 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00
VST+U49U24
VST+U49U24 - - - - - - - -
11
Mol. BioSyst., 2014
Wang X et al.
Supplementary Table S8. P-values of paired one-tailed t-tests for testing the mean difference of consistency scores between Method_1 and Method_2 based on ComBat batch effect adjustment and F-test DE analysis. The alternative hypothesis was that the mean consistency score of Method_1 was higher than that of Method_2.
Method_1 Method_2 Top_5 10 20 50 100 150 200 250VST+QN 3.6E-02 3.3E-04 3.2E-05 5.4E-07 6.5E-07 1.0E-04 1.1E-03 8.2E-03VST+RSN 5.3E-02 2.7E-04 5.0E-03 3.7E-06 2.5E-06 3.9E-04 5.4E-03 2.5E-02VST+SSN 7.2E-01 4.1E-02 3.0E-01 3.5E-01 5.5E-02 8.1E-02 2.0E-01 3.0E-01VST+VSN 7.1E-02 4.6E-05 3.5E-10 1.5E-12 5.6E-12 3.5E-09 1.8E-07 1.1E-05VST+RIN 8.3E-01 3.3E-01 9.1E-01 9.6E-01 7.8E-01 7.4E-01 6.4E-01 6.6E-01VST+Loess 6.0E-01 6.3E-03 5.3E-06 5.0E-06 1.1E-06 4.6E-04 2.7E-03 1.8E-02Log+U49 2.3E-01 1.0E-02 1.9E-04 1.2E-05 9.7E-08 3.2E-05 6.3E-04 1.0E-02Log+U24 9.5E-01 2.6E-01 5.2E-02 6.4E-03 1.2E-02 2.5E-02 6.8E-02 2.0E-01Log+U49U24 7.1E-01 3.1E-02 4.1E-03 1.5E-04 8.2E-06 3.7E-04 7.0E-04 2.2E-03VST+U49 - - - - - - - -VST+U24 9.4E-01 9.4E-01 9.9E-01 9.9E-01 9.9E-01 9.9E-01 1.0E+00 9.9E-01
VST+U49
VST+U49U24 6.5E-01 6.3E-01 7.4E-01 8.4E-01 9.3E-01 9.7E-01 9.6E-01 9.4E-01VST+QN 6.9E-04 1.8E-06 2.5E-08 1.1E-10 1.2E-09 2.3E-08 7.5E-07 3.0E-05VST+RSN 2.1E-03 1.1E-07 2.7E-06 1.8E-09 3.7E-09 2.0E-07 2.1E-06 6.4E-05VST+SSN 1.8E-01 1.5E-03 3.5E-03 1.8E-02 1.1E-03 3.5E-04 9.9E-04 6.1E-03VST+VSN 4.1E-03 1.3E-07 1.0E-14 3.8E-16 1.3E-15 2.5E-14 8.7E-12 5.3E-09VST+RIN 2.9E-01 2.8E-02 2.7E-01 5.6E-01 2.1E-01 8.4E-02 3.7E-02 7.8E-02VST+Loess 7.9E-02 1.5E-04 9.7E-10 8.8E-10 3.4E-09 2.8E-07 2.2E-06 1.1E-04Log+U49 1.2E-02 3.0E-04 1.7E-08 2.2E-09 3.2E-10 9.7E-09 2.5E-07 2.4E-05Log+U24 5.5E-01 9.3E-03 3.8E-05 4.9E-06 9.0E-06 8.3E-06 6.9E-05 2.6E-03Log+U49U24 1.3E-01 3.6E-04 9.5E-08 7.1E-09 2.5E-10 4.6E-09 1.3E-08 6.7E-07VST+U49 5.8E-02 5.6E-02 6.6E-03 1.3E-02 1.4E-02 5.2E-03 2.7E-03 1.1E-02VST+U24 - - - - - - - -
VST+U24
VST+U49U24 4.3E-02 4.1E-02 1.1E-02 5.0E-03 1.9E-02 1.2E-02 2.4E-03 1.5E-02VST+QN 1.4E-02 7.7E-05 1.5E-05 4.3E-08 7.3E-08 2.0E-06 1.1E-04 1.2E-03VST+RSN 3.5E-02 3.3E-05 1.4E-03 4.0E-07 2.7E-07 1.3E-05 4.6E-04 3.4E-03VST+SSN 6.0E-01 2.7E-02 1.6E-01 1.6E-01 1.4E-02 7.5E-03 3.7E-02 7.6E-02VST+VSN 5.0E-02 6.0E-05 1.6E-10 5.3E-14 4.2E-13 1.7E-11 1.6E-08 1.3E-06VST+RIN 7.4E-01 2.5E-01 8.1E-01 8.8E-01 5.3E-01 3.5E-01 2.7E-01 3.1E-01VST+Loess 4.6E-01 1.8E-03 5.2E-06 2.5E-07 2.6E-07 1.5E-05 2.7E-04 3.0E-03Log+U49 1.3E-01 7.6E-03 2.4E-05 3.4E-07 2.7E-08 8.8E-07 5.8E-05 1.4E-03Log+U24 9.2E-01 1.5E-01 2.5E-02 7.5E-04 1.1E-03 7.3E-04 7.5E-03 4.5E-02Log+U49U24 5.7E-01 9.2E-03 1.0E-03 3.2E-06 1.3E-07 1.3E-06 1.4E-05 1.3E-04VST+U49 3.5E-01 3.7E-01 2.6E-01 1.6E-01 7.4E-02 2.9E-02 3.9E-02 5.8E-02VST+U24 9.6E-01 9.6E-01 9.9E-01 9.9E-01 9.8E-01 9.9E-01 1.0E+00 9.9E-01
VST+U49U24
VST+U49U24 - - - - - - - -
12
Mol. BioSyst., 2014
Wang X et al.
Supplementary Table S9. P-values of paired one-tailed t-tests for testing the mean difference of consistency scores between Method_1 and Method_2 based on ComBat batch effect adjustment and limma DE analysis. The alternative hypothesis was that the mean consistency score of Method_1 was higher than that of Method_2.
Method_1 Method_2 Top_5 10 20 50 100 150 200 250VST+QN 1.3E-02 3.5E-03 2.8E-05 2.3E-07 2.8E-06 1.3E-04 1.6E-03 1.1E-02VST+RSN 4.4E-02 3.8E-03 1.3E-03 3.5E-06 1.8E-05 2.5E-04 6.0E-03 3.3E-02VST+SSN 5.0E-01 2.0E-01 2.5E-01 2.3E-01 9.0E-02 7.9E-02 2.0E-01 3.3E-01VST+VSN 8.7E-02 8.2E-04 4.8E-09 8.1E-15 1.1E-12 4.1E-11 8.5E-09 3.9E-07VST+RIN 7.5E-01 9.2E-01 9.0E-01 9.8E-01 9.6E-01 8.7E-01 8.2E-01 8.1E-01VST+Loess 1.5E-01 4.7E-02 4.6E-06 3.3E-06 3.4E-06 2.6E-04 1.6E-03 1.8E-02Log+U49 1.4E-01 7.1E-03 1.5E-06 1.3E-07 2.6E-08 5.1E-07 9.7E-05 1.1E-03Log+U24 8.0E-01 3.5E-01 7.7E-03 4.7E-03 2.7E-02 5.2E-02 1.4E-01 3.8E-01Log+U49U24 6.2E-01 1.1E-01 2.5E-04 1.5E-04 4.3E-05 5.9E-04 1.4E-03 5.0E-03VST+U49 - - - - - - - -VST+U24 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00
VST+U49
VST+U49U24 8.3E-01 7.4E-01 5.5E-01 9.4E-01 9.9E-01 9.9E-01 9.9E-01 9.9E-01VST+QN 8.8E-07 2.9E-07 4.0E-09 6.8E-16 2.9E-15 2.4E-11 1.3E-09 1.6E-07VST+RSN 2.6E-05 6.6E-08 4.0E-08 3.9E-15 3.8E-14 9.6E-11 5.2E-09 3.4E-07VST+SSN 5.0E-03 1.5E-04 7.3E-04 1.8E-06 1.5E-07 8.2E-07 5.2E-06 6.6E-05VST+VSN 1.2E-04 9.3E-09 4.0E-16 2.9E-25 8.9E-23 3.0E-19 1.9E-16 8.8E-14VST+RIN 1.6E-02 7.5E-02 1.2E-01 7.6E-02 1.2E-02 8.9E-03 5.8E-03 8.5E-03VST+Loess 7.9E-05 2.9E-05 1.4E-10 3.2E-15 2.0E-15 1.3E-10 2.1E-09 4.1E-07Log+U49 9.9E-05 3.1E-06 1.5E-10 7.5E-18 8.7E-17 4.8E-14 1.5E-11 3.1E-09Log+U24 3.3E-02 1.2E-04 7.7E-08 4.7E-11 2.8E-10 5.4E-08 1.2E-06 1.8E-04Log+U49U24 4.5E-03 1.9E-05 7.4E-10 1.0E-14 9.1E-16 5.5E-12 2.7E-11 3.0E-09VST+U49 2.6E-03 1.4E-03 6.6E-04 2.8E-06 1.9E-07 4.4E-06 4.1E-06 3.0E-05VST+U24 - - - - - - - -
VST+U24
VST+U49U24 1.6E-03 5.8E-04 1.6E-05 6.1E-08 1.1E-08 1.4E-06 1.5E-07 2.2E-06VST+QN 8.1E-04 9.4E-04 9.1E-05 1.9E-09 1.1E-08 1.3E-06 4.4E-05 4.4E-04VST+RSN 7.8E-03 5.1E-04 1.3E-03 5.7E-08 1.0E-07 3.1E-06 1.7E-04 1.3E-03VST+SSN 2.4E-01 8.9E-02 2.4E-01 4.0E-02 4.5E-03 3.5E-03 1.8E-02 4.0E-02VST+VSN 2.8E-02 3.1E-04 1.4E-09 3.4E-17 3.4E-16 1.6E-13 1.4E-10 6.8E-09VST+RIN 4.9E-01 8.1E-01 8.7E-01 8.9E-01 6.8E-01 4.2E-01 3.6E-01 3.4E-01VST+Loess 2.7E-02 2.0E-02 2.0E-05 3.4E-08 2.5E-08 3.3E-06 5.2E-05 8.3E-04Log+U49 3.6E-02 3.7E-03 3.3E-06 4.1E-10 1.2E-10 4.0E-09 1.6E-06 2.3E-05Log+U24 5.8E-01 1.7E-01 6.9E-03 1.5E-04 3.9E-04 1.3E-03 9.7E-03 6.6E-02Log+U49U24 3.2E-01 3.2E-02 1.8E-04 4.4E-07 2.2E-08 1.3E-06 9.4E-06 7.0E-05VST+U49 1.7E-01 2.6E-01 4.5E-01 5.8E-02 7.1E-03 7.9E-03 9.7E-03 1.3E-02VST+U24 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00 1.0E+00
VST+U49U24
VST+U49U24 - - - - - - - -
13
Mol. BioSyst., 2014
Wang X et al.
Supplementary Table S10. Overlapped miRNAs of the four shortlists from array-matrix split data subset. The normalization method was VST+U24, and the DE analysis method was limma.
Top_5 Top_10 Top_20 Top_50 Top_100 Top_150 Top_200 Top_250- - - - hsa-miR-409-3p hsa-miR-409-3p hsa-miR-409-3p hsa-miR-409-3p
hsa-miR-370 hsa-miR-370 hsa-miR-370 hsa-miR-370hsa-miR-22 hsa-miR-22 hsa-miR-22hsa-miR-495 hsa-miR-495 hsa-miR-495
hsa-miR-337-3p hsa-miR-337-3phsa-miR-329 hsa-miR-329hsa-miR-503 hsa-miR-503hsa-miR-431 hsa-miR-431hsa-miR-335* hsa-miR-335*hsa-miR-923 hsa-miR-923
hsa-miR-376chsa-miR-130ahsa-miR-376b
hsa-miR-664
14