Classification of data from a .csv file.

I have a csv file which contains some values which are both infected non infected data, and I want to select top 20 infected features from it. How can I classify and get the top 20 feature from the infected data.

[Star Strider]

I would use principal component analysis (the pca function) to determine the features that are most likely to distinguish between the infected and non-infected groups.

You have not provided enough information for a specific Answer, so I am listing this as a Comment.