gdcCorPlot seems to remove SolidTissueNormal samples

[liyezhou]

When I try to run the following code taken from the help passage of gdcCorPlot,

genes <- c('ENSG00000000938','ENSG00000000971','ENSG00000001036',
           'ENSG00000001084','ENSG00000001167','ENSG00000001460')

samples <- c('TCGA-2F-A9KO-01', 'TCGA-2F-A9KP-01', 
             'TCGA-2F-A9KQ-01', 'TCGA-2F-A9KR-11', 
             'TCGA-2F-A9KT-11', 'TCGA-2F-A9KW-11')

metaMatrix <- data.frame(sample_type=rep(c('PrimaryTumor',
                                           'SolidTissueNormal'),each=3),
                         sample=samples,
                         days_to_death=seq(100,600,100),
                         days_to_last_follow_up=rep(NA,6))

rnaExpr <- matrix(c(2.7,7.0,4.9,6.9,4.6,2.5,
                    0.5,2.5,5.7,6.5,4.9,3.8,
                    2.1,2.9,5.9,5.7,4.5,3.5,
                    2.7,5.9,4.5,5.8,5.2,3.0,
                    2.5,2.2,5.3,4.4,4.4,2.9,
                    2.4,3.8,6.2,3.8,3.8,4.2),6,6)
rownames(rnaExpr) <- genes
colnames(rnaExpr) <- samples
gdcCorPlot(gene1 = 'ENSG00000000938', 
           gene2    = 'ENSG00000001084',
           rna.expr = rnaExpr,
           metadata = metaMatrix)

I get the following error message

`geom_smooth()` using formula 'y ~ x'
Warning messages:
1: Use of `corDa$lncDa` is discouraged. Use `lncDa` instead. 
2: Use of `corDa$pcDa` is discouraged. Use `pcDa` instead. 
3: Use of `corDa$lncDa` is discouraged. Use `lncDa` instead. 
4: Use of `corDa$pcDa` is discouraged. Use `pcDa` instead. 
5: Removed 3 rows containing missing values (geom_point). 

and the output is as follows, which seem to have remove all three samples corresponding to SolidTissueNormal.

[liyezhou]

I've read through your source code and It seems the problem is that in the ggplot function where

scale_colour_manual(breaks = sampleType, 
                      values = c('chocolate1', 'blue')) +

should rather be

scale_colour_manual(breaks = levels(sampleType), 
                      values = c('chocolate1', 'blue')) +