I have a data file with 95 columns and many rows (over 14000). I have calculated the relative abundance and now want to verify that sum of each column is 1. But it does not work on R. I tried to test if sum of each column is real 1 (or near 1) and did it manually in excel and it did, the sum of columns was 1. I tried to change the tolerance if the issue is because of rounding numbers, but it was not. here is my code:
tolerance <- 1e-10
sums_relative_abundance <- colSums(relative_abundances)
all(abs(sums_relative_abundance - 1.0) < tolerance)
my data has 95 columns.
head(relative_abundances)
X1003 X1004 X1027 X1049
1 2.334938e-03 8.367276e-04 8.928571e-03 3.050125e-02
2 2.880184e-04 1.224950e-04 6.647746e-04 3.065996e-04
3 1.457391e-03 7.153990e-04 1.106922e-03 5.461621e-04
4 5.188696e-05 3.413513e-05 8.658196e-05 1.800099e-05
5 2.164549e-05 9.000495e-06 1.882176e-05 9.798542e-06
6 1.882176e-05 1.959708e-05 5.381118e-05 1.104796e-04
dput(head(sums_relative_abundance))
c(X1003 = 0.523067970544805, X1004 = 0.462073648305777, X1027 = 0.751016736292847,
X1049 = 1.14795999147871, X1053 = 0.892912676398911, X1058 = 1.03526073508692
)