There are many software for the calculation of molecular descriptors that can be used for free. Here a non-exhaustive list:

- edragon (http://www.vcclab.org/lab/edragon/) based on dragonX 1.4 version. It is the freeware version of the previous version of Dragon software (https://chm.kode-solutions.net/products_dragon.php), one of the most known software for the calculation of molecular descriptors.

- ISIDA fragmentor (http://infochim.u-strasbg.fr/spip.php?rubrique49), it calculates just Substructural Molecular Fragments (SMF) as well as ISIDA Property-Labelled Fragments (IPLF) descriptors.

- PaDeL (http://padel.nus.edu.sg/software/padeldescriptor/) based on CDK.

You can find a more detailed list of freeware and commercial software for the calculation of molecular descriptors here:

http://www.moleculardescriptors.eu/softwares/softwares.htm

There are many software for the calculation of molecular descriptors that can be used for free. Here a non-exhaustive list:

- edragon (http://www.vcclab.org/lab/edragon/) based on dragonX 1.4 version. It is the freeware version of the previous version of Dragon software (https://chm.kode-solutions.net/products_dragon.php), one of the most known software for the calculation of molecular descriptors.

- ISIDA fragmentor (http://infochim.u-strasbg.fr/spip.php?rubrique49), it calculates just Substructural Molecular Fragments (SMF) as well as ISIDA Property-Labelled Fragments (IPLF) descriptors.

- PaDeL (http://padel.nus.edu.sg/software/padeldescriptor/) based on CDK.

You can find a more detailed list of freeware and commercial software for the calculation of molecular descriptors here:

http://www.moleculardescriptors.eu/softwares/softwares.htm

preadmet an online tool

In addition to Andrea Mauri, you also have:

- Molecular Operating Environment (MOE, non-free);

- MarvinSketch (http://www.chemaxon.com, freeware).

Best regards.

I reported just a limited list of free software.

MarvinSketch is free only for academic. On the moleculardescriptors.eu website there is a more complete list.

Obviously there are also more commercial software:

- Dragon by Talete srl - Prof. Roberto Todeschini (https://chm.kode-solutions.net/products_dragon.php);

- CODESSA by Semichem - Prof. Alan Katritzky (http://www.semichem.com/codessa/default.php);

- Adriana code by molecular networks - Prof. Johann Gasteiger (http://www.molecular-networks.com/products/adrianacode);

- Sarchitect (http://www.strandls.com/sarchitect/index.html)

....

@Deepak yeah it is good but i am unabe to export the result to .XLS and CVS file...

I use schrodinger suite for all the insilico work. I just have an idea what preadmet can do. You can try schrodinger suite for better and accurate results. And all the data can be saved in desired file formats

I use Vlife MDS to calculate molecular descriptors, it can calculate descriptors for the whole molecule and fragment-based descriptors

It very much depends what type of descriptors you wish to calculate as well as your resources. Do you have any idea (based upon a mechanistic understanding) of which descriptors are most appropriate for your problem?

If you are just trying to calculate as many descriptors as possible, I can suggest the following software programs.

Commercial

Molecular Operating Environment from the Chemical Computing Group

Pipeline Pilot from Accelrys - computes extended connectivity fingerprints (ECFP, FCFP types) which you could convert into a count vector/bit vector using a script or - possibly -using a component within the software.

This also computes some other topological indices, P_VSA descriptors, logD, logP....

DRAGON - will give you thousands of descriptors.

There is also a limited online version of the software: E-DRAGON.

Free for Academics

ChemAxon's (www.chemaxon.com) software tools offer a variety of different kinds of descriptors.

Open source freeware

OpenBabel will calculate some descriptors and fingerprints

CDK

RDKit?

jCompoundMapper will compute various fingerprints and output these in various file formats which may be suitable for your modelling/similarity searching software

HTH

Richard

I have a problem with PreADMET. I can't open the program for drawing chemical structures, so I can't calculate any descriptor that I need. Could you tell me if I have to install some add-on for Internet explorer to run this program? Thank you in advance.

See also QuBiLs-MIDAS (from ToMoCoMD-CARDD) at www.tomocomd.com

see the last paper (free) about comparison with all 3-4D methods by usign several datasets. http://jcheminf.springeropen.com/articles/10.1186/s13321-016-0122-x

We have developed two free platform to calculate various molecular descriptors (more than 3000 descriptors) and fingerprints. http://www.scbdd.com/chemdes

http://biotriangle.scbdd.com

Have a look at alvaDesc:

https://www.alvascience.com/alvadesc/

It calculates a comprehensive collection of more than 5000 molecular descriptors (0D, 1D, 2D, 3D molecular descriptors). It can also calculate both MACCS166 fingerprint and a customisable version of hashed molecular fingerprint (Extended Connectivity Fingerprint and Path Fingerprint).

alvaDesc can be used online as a service for non commercial purposes at OCHEM

https://ochem.eu/descriptorscalculator/show.do

For those who have some basic experience of Python or are willing to acquire basic Python notions:

RDkit is a quick and free way to get a bunch of descriptors, which range from 1D to 3D. Also, note that if your molecular names are not completely niche, you can easily convert them into SMILES using the cirpy library. I had a good experience with this library: from a bunch of about 4000 molecules, it could convert name to smiles automatically for about 3800 of them and I had just 200 to convert manually. Once you have SMILES of your molecules, descriptors can be easily calculated with RDkit.

Check the list of descriptors, 3D descriptors and fingerprints here:

https://www.rdkit.org/docs/GettingStartedInPython.html#list-of-available-descriptors

Note that fragment descriptors, for group contributions, can also be computed once suitably defined.

I recommend using the last version of Python as installed with Anaconda. This way, you will have the least possible hassle with adding new libraries. Personally I simply use notepad++ as the text editor and I execute my script using the windows command line. This way, the entire workflow is free.

As an example, here is a simple script I've done that converts a bunch of names (1 by line) given in names.txt file into their SMILES and prints them (as well as raw formula) on the command line output.

#!/usr/bin/python

# -*- coding: Latin-1 -*-

# ----- Get SMILES from list of names -----

#

#

# (c) 2019, University of Nanjing

#

# A: Theophile Gaudin

#

import cirpy

with open("names.txt") as file:

data = file.readlines()

for molecule in data:

try:

name = molecule.rstrip("\n\r")

SMILES = cirpy.resolve(name, 'smiles')

formula = cirpy.resolve(SMILES, 'formula')

Mw = cirpy.resolve(SMILES, 'Mw')

print(name, ";", SMILES, ";", formula, ";", Mw)

except:

print("none")

What I have done is crosscheck results from cirpy with the ones from pubchem because I have found that to get SMILES from names sometimes cirpy is not as accurate. I usually get it with the pubchempy module and do some crosschecking with a second source. Also with a normal webrequest and one regex you can also use the NIST webbook to search by name. Which is by far the most reliable source I have found to get SMILES from names.

Maybe this helps aswell:

https://github.com/sabidhasan/NistPy

Also if you want you could try KNIME. It gives tools for data retrieval, comparison and filtering.

In addition to Simon Müller answer I suggest CIR for KNIME:

https://www.alvascience.com/knime-cir/

https://www.knime.com/book/chemical-identifier-resolver-for-knime-trusted-extension

https://nodepit.com/node/com.alvascience.cir.CIRNodeFactory

You can use Padel, it is freely available, and i am having the license for Dragon 7, if you can sgare your files in sdf format I will calculate descriptors for you.