Simon

My algorithm is not generally tailored for UNIFAC. UNIFAC was just used as example. There are two algorithms in my paper. The simple one would require some tweaking to work...

The idea about using NLP sounds quite interesting. I don't know much about NLP and had never thought of using it although I thought of the idea of automagically generating...

@tyyfkpo You can learn how to define SMARTS [here](https://www.daylight.com/dayhtml/doc/theory/theory.smarts.html). I can also recommend [this webpage](https://smarts.plus/) to test your SMARTS. For simple things you will be able to use GetSubstructMatch. For...

Yes definetly, the code was developed to be used with whatever SMARTS groups you need. I recommend the official [SMARTS documentation ](https://www.daylight.com/dayhtml/doc/theory/theory.smarts.html)and having a look at [smarts.plus](https://smarts.plus) when you are...

> Yes definetly, the code was developed to be used with whatever SMARTS groups you need. I recommend the official [SMARTS documentation ](https://www.daylight.com/dayhtml/doc/theory/theory.smarts.html)and having a look at [smarts.plus](https://smarts.plus) when you...

> Hi @simonmb . I read your paper on this issue. I found a list of 119 groups and their SMILES in your paper. Is is possible to extend these...

The error you describe is correct, as a single aromatic atom is not a molecule you can fragment. You probably have the 10000 molecules from somewhere and there might be...

@sardaar11 please open an issue in the [fragmenter repository](https://github.com/simonmb/fragmentation_algorithm) and provide the details. i.e. example code to reproduce the error.

Wow, thank you for the quick response! Yes, I can confirm it is working again. Was this problem already known?

Hi, this time it was not me, but the server is offline. Or is this on purpose?