ugropy is a Python library to obtain subgroups from different thermodynamic group contribution models using both the name or the SMILES representation of a molecule. If the name is given, the library uses the PubChemPy library to obtain the SMILES representation from PubChem. In both cases, ugropy uses the RDKit library to search the functional groups in the molecule.

ugropy is tested for Python 3.10, 3.11, 3.12, and 3.13 on Linux, Windows and Mac OS.

You can access the documentation here: https://ipqa-research.github.io/ugropy/

Try ugropy now

You can try ugropy without installing it by clicking on the Colab badge.

You can install ugropy by:

pip install ugropy

Citing ugropy

ugropy now has an article! If you use ugropy in your research, please cite:

@article{brandolin2025ugropy,
  title={Ugropy: An Extensible Python Package for Thermodynamic Model Functional Group Identification via Mathematical Optimization},
  author={Brandol{\'\i}n, Salvador E and Benelli, Federico E and Magario, Ivana and Scilipoti, Jos{\'e} A},
  journal={Industrial \& Engineering Chemistry Research},
  volume={64},
  number={35},
  pages={17217--17227},
  year={2025},
  publisher={ACS Publications},
  doi = {10.1021/acs.iecr.5c02552}
}

Check the publication here.

Models implemented

Gibbs / EoS models

• Classic liquid-vapor UNIFAC
• Predictive Soave-Redlich-Kwong (PSRK)
• Dortmund (modified UNIFAC)

Property estimators

• Joback
• Abdulelah-Gani (beta)

Writers

ugropy allows you to convert the obtained functional groups or estimated properties to the input format required by the following thermodynamic libraries:

• Clapeyron.jl
• Thermo
• yaeos (Fortran)

Example of use

Here is a little taste of ugropy, please, check the full tutorial here to see all it has to offer!

Get groups from the molecule's name:

from ugropy import Groups


hexane = Groups("hexane")

print(hexane.unifac.subgroups)
print(hexane.psrk.subgroups)
print(hexane.dortmund.subgroups)
print(hexane.joback.subgroups)
print(hexane.agani.primary.subgroups)

{'CH3': 2, 'CH2': 4}
{'CH3': 2, 'CH2': 4}
{'CH3': 2, 'CH2': 4}
{'-CH3': 2, '-CH2-': 4}
{'CH3': 2, 'CH2': 4}

Get groups from molecule's SMILES:

propanol = Groups("CCCO", "smiles")

print(propanol.unifac.subgroups)
print(propanol.psrk.subgroups)
print(propanol.dortmund.subgroups)
print(propanol.joback.subgroups)
print(propanol.agani.primary.subgroups)

{'CH3': 1, 'CH2': 2, 'OH': 1}
{'CH3': 1, 'CH2': 2, 'OH': 1}
{'CH3': 1, 'CH2': 2, 'OH (P)': 1}
{'-CH3': 1, '-CH2-': 2, '-OH (alcohol)': 1}
{'CH3': 1, 'CH2': 2, 'OH': 1}

Estimate properties with the Joback and Abdulelah-Gani models!

limonene = Groups("limonene")

print(limonene.joback.subgroups)
print(f"{limonene.joback.critical_temperature} K")
print(f"{limonene.joback.vapor_pressure(176 + 273.15)} bar")

{'-CH3': 2, '=CH2': 1, '=C<': 1, 'ring-CH2-': 3, 'ring>CH-': 1, 'ring=CH-': 1, 'ring=C<': 1}
657.4486692170663 kelvin
1.0254019428522743 bar

print(limonene.agani.primary.subgroups)
print(limonene.agani.secondary.subgroups)
print(limonene.agani.tertiary.subgroups)
print(f"{limonene.agani.critical_temperature}")
print(limonene.agani.molecular_weight / limonene.agani.liquid_molar_volume)

{'CH3': 2, 'CH2=C': 1, 'CH2 (cyclic)': 3, 'CH (cyclic)': 1, 'CH=C (cyclic)': 1}
{'CH3-CHm=CHn (m,n in 0..2)': 1, '(CHn=C)cyc-CH3 (n in 0..2)': 1, 'CHcyc-C=CHn (n in 1..2)': 1}
{}
640.1457030826214 kelvin
834.8700605718585 gram / liter

Visualize your results! (The next code creates the ugropy logo)

mol = Groups("CCCC1=C(COC(C)(C)COC(=O)OCC)C=C(CC2=CC=CC=C2)C=C1", "smiles")

mol.unifac.draw(
    title="ugropy",
    width=800,
    height=450,
    title_font_size=50,
    legend_font_size=14
)

ugropy can obtain multiple solutions, even nonoptimal ones if desired. For example:

from ugropy import unifac


solutions = unifac.get_groups(
    "9,10-dihydroanthracene",
    search_multiple_solutions=True,
    search_nonoptimal=True
)

for sol in solutions:
    print(sol.subgroups)

{'ACH': 8, 'AC': 2, 'ACCH2': 2}
{'CH2': 1, 'ACH': 8, 'AC': 3, 'ACCH2': 1}
{'CH2': 2, 'ACH': 8, 'AC': 4}

Write down the Clapeyron.jl .csv input files.

from ugropy import writers

names = ["limonene", "adrenaline", "Trinitrotoluene"]

grps = [Groups(n) for n in names]

# Write the csv files into a database directory
writers.to_clapeyron(
    molecules_names=names,
    unifac_groups=[g.unifac.subgroups for g in grps],
    psrk_groups=[g.psrk.subgroups for g in grps],
    joback_objects=[g.joback for g in grps],
    path="database"
)

Obtain the Caleb Bell's Thermo subgroups

from ugropy import unifac

names = ["hexane", "ethanol"]

grps = [unifac.get_groups(n) for n in names]

[writers.to_thermo(g.subgroups, unifac) for g in grps]

[{1: 2, 2: 4}, {1: 1, 2: 1, 14: 1}]