1
$\begingroup$

This is one of the first steps to be done while analysing chemical data or applying training models to predict chemical activity. However, I am unclear as to why is this done. Doe sit have a major effect on the result if duplicates are not removed? What is the main reason why we remove duplicate compounds?

Thanks

$\endgroup$

1 Answer 1

1
$\begingroup$

Let's consider a data set with three attributes for each compound: a chemical name, a CAS number (unique identifier for a compound) and a single empirically measured numerical interest, such as efficacy in lowering blood pressure in rats. If you had duplicate entries for a single experimental result, that could potentially bias the analysis. If you had two results for the same compound from two different experiments or measurements, you would surely want to include both in the analysis. If you had two results for the same compound (same CAS number) but named differently, you would want to keep them both in your data set, but use only the CAS number as the nominal variable for structure.

$\endgroup$
1
  • $\begingroup$ In any case it is well worth going over the data set with a fine-toothed comb to ensure that it is internally consistent, before running an analysis. $\endgroup$
    – iad22agp
    Jul 14, 2020 at 12:53

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.