💻 CHEM 281 Lab 7: Molecules, Templates, Lambdas, Rust

🧪 Goal

The goal of this lab is to:

1. Familiarize yourself with representing chemical structures in a computer.
2. Practice using C++ templates and lambdas.
3. Learn how to write and compile Rust code.
4. Work with 3D molecular coordinates generated from SMILES strings.

---

🗂️ Provided

• Starter code in C++ and Rust to read and store molecules.
• Helper functions
• Docker environment with rdkit.
• Python script to create mol files: scripts/smi_to_mol.py

---

💻 Setup

./build_image.sh # You may need to chmod +x
./interactive.sh 

✅ Tasks

1. Generate a .mol2 file

Use the provided script to generate .mol files:

python scripts/smi_to_mol.py

Then once its made you should have a new directory with all the .mol files files/mols. You need to then use openbabel to convert all the mol files into a single .mol2 file. If you change the name of any of these files then the executable based on main.cpp won't be able to find it so you can just update the names then.

obabel -imol files/mols/*.mol -O files/mols.mol2

2. Complete the C++ functions

filter_molecules You should apply the following filters to the list of molecules to trim their size down from 80. Use the templated function filter() found in molecule.h to help you. 1. Only molecules with atoms <= 25 are allowed. 2. Only molecules with ALL single bonds are allowed. 3. Only molecules with more than 1 HBA and 1 HBD are allowed.

These filters are excluding molecules that are too large (lets pretend our active site is really small!), and contains molecules that have a lot of degrees of freedom. For 3) you will need to complete the following functions:

countHBondDonors A HBond donor for our purposes is an Oxygen or Nitrogen atom with a full valency with atleast 1 hydrogen attached. The way we would look for them is by checking if the total bond order of an Oxygen is 2 and it has 1 hydrogen as a neighbor. For Nitrogens the total bond order would be 3 and it has 1 hydrogen as a neighbor.

countHBondAcceptors A HBond acceptor for our purposes is an Oxygen or Nitrogen atom with a full valency. The way we would look for them is by checking if the total bond order of an Oxygen is 2. For Nitrogens the total bond order would be 3.

Note: A double bonded Oxygen has a bond order of 2. Note: An atom can be considered both an acceptor and a donor. Example: Water oxygen.

Build and run the executable ./mol_exe from the build.

mkdir build
cd build
make
./mol_exe

Running ./mol_exe

root@dfa54b69bdf3:/repo/build# ./mol_exe
80
28
H
0.8257
2.0227
-1.1458
Small molecules: 36
Single bonds: 34
Filtered mols: 13
HBond donors: 19
HBond acceptors: 56

3. Complete the Rust functions

Complete count_hbond_acceptors and count_hbond_donors in molecule.rs. They are identical to the C++ functions just written in Rust! Once completed, filter the molecules but instead of using an explicit filter_molecules function, chain iterators together to to achieve the same functionality.

Build and run:

cd molecule
cargo run

Loaded 80 molecule(s)
36 small molecule(s)
19 hbd molecule(s)
56 hba molecule(s)

Extra time

Spend some time looking at the differences and similarities between the C++ code and the Rust code. You can also look at the smi_to_mol.py script which was used to generate 3D structures from SMILES.