RDKIT descriptors

I have calculated descriptors using “RDKIT descriptor calculation” node, but it generated only ~127 descriptors where in reality there are around 195 descriptors by RDKIT, so any idea how i can get the full list of descriptors calculated by RDKIT?

Thank you

I don’t think there’s enough information here for us to be able to really help.

Which specific descriptors are you “missing”? Also, where are you getting these totals from?

Depending on which descriptor(s) you want, there may be other nodes that provide this functionality.
Alternatively you can use a Python node and use the RDKit directly.

Taking ChemDes as an example, it shows 196 descriptors in the RDKit library
However using “RDKIT descriptor calculation” node, only ~127 descriptors are found in the output. I haven’t noted the missing/additional descriptors yet, however there are descriptors not calculated by just using this KNIME node mentioned, so how can I get a full set/maximum number of descriptors using RDKit calculated through KNIME nodes?

You might need to use RDKit in Python to calculate the other descriptors. You could save those to a csv file and then merge them to the ones you get from KNIME.

1 Like

We’re just going in circles.

The RDKit only lists 145 1D and 2D descriptors. There are indeed differences between this list and the descriptors available in the RDkit Descriptor Calculation node; I get 119 descriptors being calculated. Neither of these numbers matches your output. You’ll need to investigate further to figure out what else ChemDes is including.

Like I said in my last response, if you know what descriptor(s) you need that’s missing, there may be other nodes (there are a few of them) that provide this functionality, or you can use a Python node to use the RDKit (or some other package) bring it in. Without further details from you, that’s all that can be said.

1 Like