#!/usr/bin/env python
# coding: utf-8

# **An Introduction to the RDKit for Cheminformatics**

# Import the necessary Python libraries

# In[1]:


from rdkit import Chem #RDKit Chemistry
from rdkit.Chem.Draw import IPythonConsole #RDKit drawing
from rdkit.Chem import Draw #RDKit drawing
# A few settings to improve the quality of structures 
from rdkit.Chem import rdDepictor
IPythonConsole.ipython_useSVG = True
rdDepictor.SetPreferCoordGen(True)
from rdkit.Chem import PandasTools #Add the ability to add a molecule to a dataframegrid
import mols2grid #The mols2grid library provides a convenient way of displaying molecules in a grid


# Create a molecule (benzene) from a SMILES string

# In[2]:


mol = Chem.MolFromSmiles("c1ccccc1")


# In[3]:


mol


# Get SMILES fro Gleevec from ChEMBL
# 
# - [From ChEMBL](https://www.ebi.ac.uk/chembl/compound/inspect/CHEMBL941)

# In[4]:


glvc = Chem.MolFromSmiles("CN1CCN(Cc2ccc(cc2)C(=O)Nc3ccc(C)c(Nc4nccc(n4)c5cccnc5)c3)CC1")


# In[5]:


glvc


# Read a set of molecules from an SD file

# In[6]:


mols = [x for x in Chem.SDMolSupplier("example_compounds.sdf")]


# In[7]:


mols


# Draw these molecules as a grid

# In[8]:


Draw.MolsToGridImage(mols,molsPerRow=4,useSVG=True)


# We can use the mols2grid library to display molecules in a grid

# In[9]:


mols2grid.display(mols)


# In[10]:


mols2grid.selection


# We can also read an SD file into a Pandas dataframe. 

# In[11]:


df = PandasTools.LoadSDF("example_compounds.sdf")


# In[12]:


df.head()


# Let's add columns with molecular weight and LogP to the dataframe.

# In[13]:


from rdkit.Chem.Descriptors import MolWt
from rdkit.Chem.Crippen import MolLogP
df['MW'] = [MolWt(x) for x in df.ROMol]
df['LogP'] = [MolLogP(x) for x in df.ROMol]


# In[14]:


df.head()


# We can use a boxplot to examine the distribution of molecular weight within the dataframe. 

# In[15]:


import seaborn as sns


# In[16]:


ax = sns.boxplot(x=df.MW)


# In[ ]: