Skip to main content

Drug Development

More on “Metabolite Likeness” as a Predictor

A recent computational paper that suggested that similarity to known metabolites could help predict successful drug candidates brought in a lot of comments around here. Now the folks at Cambridge MedChem Consulting have another look at it here.
The big concern (as was expressed by some commenters here as well) is the Tanimoto similarity cutoff of 0.5. Does that make everything look too similar, or not? CMC has some numbers across different data sets, and suggests that this cutoff is, in fact, too permissive to allow for much discrimination. People with access to good comparison sets of compounds that made it and compounds that didn’t – basically, computational chemists inside large industrial drug discovery organizations – will have a better chance to see how all this holds up.

6 comments on “More on “Metabolite Likeness” as a Predictor”

  1. TX raven says:

    Ok. Let’s make it the “rule of 0.85” then.
    Or perhaps 0.9, to be on the safe side 🙂

  2. Anonymous says:

    Can we just cut to the chase and call it “metabolikeness”?

  3. Anonymous says:

    Can we just cut to the chase and call it “metabolikeness”?

  4. Cellbio says:

    We need a much more impressive name, like, the Tanimotome. The field would be Tanimotology, of course. Reflecting recent trends in discipline creep, one can look forward to the few remnant Pharmacology departments that missed the Systems Biology bandwagon to be rebranded as the Department of Translational Tanimotology.

  5. mirco says:

    It is not just the metrics used, but it also depends on which fingerprints have been used.

  6. SM says:

    I wonder if those libraries based on metabolites have more successful hit rates?

Comments are closed.