Data Summary

This is a summary of the given data matrix for the first 12 genes.

Table continues below
V1 V2 V3 V4 V5 V6
Min. : 3.0 Min. : 0.0 Min. : 0 Min. : 0 Min. : 0 Min. : 0.0
1st Qu.: 952.5 1st Qu.: 12.8 1st Qu.: 2541 1st Qu.: 1231 1st Qu.: 2982 1st Qu.: 107.2
Median : 16976.0 Median : 2723.5 Median : 18304 Median : 5343 Median : 13666 Median : 6351.5
Mean : 38512.8 Mean : 50149.8 Mean : 58634 Mean : 29463 Mean : 33469 Mean : 42320.7
3rd Qu.: 43151.0 3rd Qu.: 39270.5 3rd Qu.: 59952 3rd Qu.: 25138 3rd Qu.: 38146 3rd Qu.: 43330.8
Max. :433393.0 Max. :674276.0 Max. :825606 Max. :562960 Max. :338182 Max. :773652.0
V7 V8 V9 V10 V11 V12
Min. : 0 Min. : 0 Min. : 0.0 Min. : 0.0 Min. : 0.0 Min. : 0
1st Qu.: 3512 1st Qu.: 3320 1st Qu.: 976.5 1st Qu.: 217.8 1st Qu.: 333.8 1st Qu.: 2430
Median : 14372 Median : 14070 Median : 11117.5 Median : 10362.0 Median : 2640.5 Median : 26740
Mean : 43933 Mean : 40692 Mean : 38255.1 Mean : 58424.7 Mean : 38351.8 Mean : 62510
3rd Qu.: 54410 3rd Qu.: 56977 3rd Qu.: 47964.0 3rd Qu.: 56310.5 3rd Qu.: 20305.8 3rd Qu.: 78459
Max. :362156 Max. :238133 Max. :234333.0 Max. :780273.0 Max. :794217.0 Max. :411485

Heatmap

This is a heatmap of the given data matrix showing the batch effects and variations with different conditions.

Sample Correlations

This is a heatmap of the correlation between samples.

This plot helps identify outlying samples.

PCA: Principal Component Analysis

This is a plot of the top two principal components and showing the variation with respect to batch effects and different conditions.

PCA Proportion Variation and correlation

PCA Proportion Variation and correlation Table

Percentage Variation Cumulative Percentage Variation Condition Correlation Batch Correlation
21.3 21.3 6.66 3.61
10.78 32.08 0.04 7.22
7.63 39.71 0.49 0.81
6.68 46.39 0.03 0.84
5.79 52.18 0.09 3.89
5.49 57.67 9.1 11.76
5.04 62.71 6.9 0.69
4.43 67.14 0.93 0.02
3.85 70.99 0.45 0.2
3.34 74.33 0 3.34
3.07 77.4 0.02 3.72
2.75 80.15 0.89 4.79
2.2 82.35 0.09 0.04
1.93 84.28 0.48 0.11
1.79 86.07 1.19 9.36
1.65 87.72 1.53 2.08
1.41 89.13 0.86 1.84
1.2 90.33 0.38 2.02
1.16 91.49 0.36 0.6
0.87 92.36 4.91 0.01
0.83 93.19 8.47 0.4
0.74 93.93 0 1.9
0.72 94.65 0.08 0.08
0.65 95.3 0.44 0.8
0.56 95.86 1.72 0.03
0.55 96.41 0.14 1.66
0.47 96.88 2.51 2.34
0.4 97.28 0.43 3.44
0.36 97.64 2.2 0.5
0.32 97.96 1.95 0.28
0.28 98.24 0.12 0.14
0.26 98.5 0.01 0.72
0.22 98.72 0.02 1.42
0.2 98.92 2.07 0.02
0.19 99.11 0.98 1.18
0.17 99.28 0.7 8.36
0.14 99.42 0.44 0.99
0.12 99.54 5.25 0.9
0.1 99.64 0.29 0.25
0.08 99.72 2.06 0.06
0.05 99.77 1.51 0.81
0.05 99.82 1.6 0.53
0.04 99.86 0.06 0.16
0.03 99.89 0.01 1.01
0.03 99.92 0 1.16
0.02 99.94 0.55 1.47
0.02 99.96 0.02 0.66
0.01 99.97 3.91 0.32
0 99.97 0.66 0.51
0 99.97 5.59 0.06

Combat Plots

This is a plot showing whether parametric or non-parameteric prior is appropriate for this data. It also shows the Kolmogorov-Smirnov test comparing the parametric and non-parameteric prior distribution.

## Found 3 batches
## Adjusting for 1 covariate(s) or covariate level(s)
## Standardizing Data across genes
## Fitting L/S model and finding priors

## 
##  One-sample Kolmogorov-Smirnov test
## 
## data:  gamma.hat[1, ]
## D = 0.1321, p-value = 0.3187
## alternative hypothesis: two-sided

Batch Effects testing

This is a summary of the statistical test for batch effects.

## 
## Call:
## lm(formula = pc[, 1] ~ fbatch + fcond)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.2124 -0.9805 -0.2123  0.6178 11.7385 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)  
## (Intercept) -0.70903    0.53661  -1.321   0.1918  
## fbatch2     -0.08874    0.65721  -0.135   0.8931  
## fbatch3      1.35003    0.65721   2.054   0.0446 *
## fcond1       0.57721    0.53661   1.076   0.2867  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 2.078 on 56 degrees of freedom
## Multiple R-squared:  0.1136, Adjusted R-squared:  0.06612 
## F-statistic: 2.392 on 3 and 56 DF,  p-value: 0.07812
## 
## 
## Call:
## lm(formula = pc[, 2] ~ fbatch + fcond)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -6.5546 -0.5030  0.0587  0.7284  5.7549 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)
## (Intercept)  0.06335    0.47433   0.134    0.894
## fbatch2     -0.31186    0.58094  -0.537    0.594
## fbatch3      0.37279    0.58094   0.642    0.524
## fcond1      -0.16733    0.47433  -0.353    0.726
## 
## Residual standard error: 1.837 on 56 degrees of freedom
## Multiple R-squared:  0.02637,    Adjusted R-squared:  -0.02578 
## F-statistic: 0.5057 on 3 and 56 DF,  p-value: 0.68
## 
## 
## Call:
## lm(formula = pc[, 3] ~ fbatch + fcond)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -2.9648 -0.9155  0.0228  0.6581  5.2799 
## 
## Coefficients:
##              Estimate Std. Error t value Pr(>|t|)  
## (Intercept)  0.779111   0.430342   1.810   0.0756 .
## fbatch2     -0.793963   0.527059  -1.506   0.1376  
## fbatch3     -0.005443   0.527059  -0.010   0.9918  
## fcond1      -1.025284   0.430342  -2.382   0.0206 *
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.667 on 56 degrees of freedom
## Multiple R-squared:  0.1342, Adjusted R-squared:  0.08784 
## F-statistic: 2.894 on 3 and 56 DF,  p-value: 0.04318
## 
## 
## Call:
## lm(formula = pc[, 4] ~ fbatch + fcond)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.6322 -0.8311 -0.2285  0.4871  7.2334 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)
## (Intercept)  -0.4656     0.4426  -1.052    0.297
## fbatch2       0.2736     0.5421   0.505    0.616
## fbatch3       0.5671     0.5421   1.046    0.300
## fcond1        0.3708     0.4426   0.838    0.406
## 
## Residual standard error: 1.714 on 56 degrees of freedom
## Multiple R-squared:  0.03108,    Adjusted R-squared:  -0.02082 
## F-statistic: 0.5988 on 3 and 56 DF,  p-value: 0.6184
## 
## 
## Call:
## lm(formula = pc[, 5] ~ fbatch + fcond)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -5.7811 -0.6142  0.1258  0.6663  3.5820 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)  
## (Intercept)   0.7510     0.4236   1.773   0.0817 .
## fbatch2      -0.9589     0.5188  -1.848   0.0699 .
## fbatch3      -0.7490     0.5188  -1.444   0.1544  
## fcond1       -0.3634     0.4236  -0.858   0.3947  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.641 on 56 degrees of freedom
## Multiple R-squared:  0.07455,    Adjusted R-squared:  0.02497 
## F-statistic: 1.504 on 3 and 56 DF,  p-value: 0.2235