3

I've read similar questions about that here, but still didn't work for my problem. I have a table with: columns with different kinds of pollutants levels and a column with the name of the samples. However, in no way I can change the name of the labels (now it's numbers) to the samples names.

Here is my script:

read.csv2("PCA_ALL.csv", header=TRUE)->tabela
tabela

pca <- prcomp(~Ter+Hop+UCM+AHS+S_Chain+L_chain+Alkyl+HMW+LMW+TOC, scale = TRUE)

fviz_pca_biplot(pca, geom = c("point","text"), 
                addEllipses = TRUE, ggtheme = theme_gray(), 
                col.var = "black", repel=FALSE, 
                title = "PCA - GB", xlab="PC1 (39%)", ylab="PC2 (22%)")

Does anyone know what I could do? Thanks in advance!

The first column of my table is named "area", where the name of the samples are.

Mari
  • 43
  • 1
  • 5
  • Could you share a sample of your data so that we can test solutions and try to help out? You could sample via: `dput(head(tabela, n))` – Kris Oct 23 '19 at 18:05
  • 1
    I just share the table in the post above. Thanks in advance! – Mari Oct 23 '19 at 18:37

1 Answers1

2

Here is a possibile solution:

tabela <- read.table(text="
         area  Ter  Hop   UCM   AHS S_Chain L_chain Alkyl  HMW   LMW  TOC
1   2010-2004 0.71 3.10 119.4 136.8    0.10    3.48 11.50 7.70 16.70 1.19
2   2004-1999 0.57 2.77  71.0  89.3    0.04    2.61  3.74 3.61  1.30 0.81
3   1999-1993 0.87 3.10 117.7 132.1    0.04    2.77  6.38 3.08  1.94 0.90
4   1993-1988 0.49 2.69  98.4 111.7    0.04    2.64  4.60 3.57  1.81 0.86
5   1988-1982 1.44 4.43  80.0  93.9    0.05    3.10  7.27 3.84  2.92 1.06
6   1982-1977 0.57 4.80  55.1  65.5    0.03    3.80  9.87 5.50  8.80 1.25
7   1977-1971 0.62 3.16 174.7 190.8    0.04    3.15  6.00 3.58  1.51 1.08
8   1971-1966 1.17 5.77 162.3 174.8    0.04    3.39  5.95 5.68  2.65 1.11
9   1966-1960 1.28 8.13 155.4 194.6    0.05    5.61  5.74 5.47  2.45 1.16
10 1960-1954  0.69 6.77  96.0 134.2    0.04    5.51  3.74 4.73  2.41 1.16
11  1954-1949 0.75 4.56  65.8 122.6    0.07    4.91  7.97 5.33  2.48 0.83
12  1949-1943 0.58 4.74  70.2 112.3    0.05    5.38  4.94 7.47  3.19 1.19
13  1943-1938 3.00 6.22  66.9 105.9    0.08    5.78 16.20 8.40  3.79 1.19
14  1938-1932 0.77 3.44  96.4 141.7    0.06    4.93  8.48 3.60  4.12 1.06
15  1932-1927 0.40 4.37  84.7 126.3    0.04    4.36  3.73 3.67  1.66 0.90
16  1927-1921 1.95 5.06  51.8  92.5    0.07    5.08 11.74 4.36  5.29 1.01
", header=T)

pca <- prcomp(~Ter+Hop+UCM+AHS+S_Chain+L_chain+Alkyl+HMW+LMW+TOC, scale = TRUE, data=tabela)
# Set row names for the matrix with rotated data
dimnames(pca$x)[[1]] <- tabela$area

library(factoextra)
fviz_pca_biplot(pca, geom = c("point","text"), 
                addEllipses = TRUE, ggtheme = theme_gray(), 
                col.var = "black", repel=FALSE, 
                title = "PCA - GB", xlab="PC1 (39%)", ylab="PC2 (22%)")

enter image description here

Marco Sandri
  • 23,289
  • 7
  • 54
  • 58
  • I share my table in the post above. I tried your solution but I still need the name of the samples. Thank you for the help! – Mari Oct 23 '19 at 18:55
  • Now it worked! Thank you very much for the help, salved my life haahahahahaha – Mari Oct 24 '19 at 00:15
  • There is just a little problem, when I change the script to show all the samples (as yours shows only some o them), this message appears: Error in dimnames(pca$x)[[1]] <- tabela$area : length of 'dimnames' [1] not equal to array extent in this case, what should I do? – Mari Oct 24 '19 at 02:19