This blog post is supporting my poster at Young Modellers Forum and makes things way easier to see and understand. Underneath each GIF, is the explanation of what you should look for as things denoise throughout the diffusion trajectory. Click the GIFs for higher quality viewing!
Continue readingTag Archives: Small Molecules
Design your very own drug: An introduction to structure-based small molecule drug design
Are you curious about how scientists design small molecules to treat disease using computational tools, but the words RDKit, docking, and QED mean nothing to you? Look no further than these tutorials for learning the fundamentals of computational small molecule drug design through interactive tutorials that introduce the key tools, concepts, and workflows. From generating compounds to evaluating their drug-likeness and binding potential, by the end you’ll be ready to explore how computational methods can result in the discovery of your very own (virtual) drug candidates to cure Zika!
Find the materials here: https://github.com/oxpig/dtc-struc-bio-smolecules/tree/main.
Continue readingEstimating the Generalisability of Machine Learning Models in Drug Discovery
Machine learning (ML) has significantly advanced key computational tasks in drug discovery, including virtual screening, binding affinity prediction, protein-ligand structure prediction (co-folding), and docking. However, the extent to which these models generalise beyond their training data is often overestimated due to shortcomings in benchmarking datasets. Existing benchmarks frequently fail to account for similarities between the training and test sets, leading to inflated performance estimates. This issue is particularly pronounced in tasks where models tend to memorise training examples rather than learning generalisable biophysical principles. The figure below demonstrates two examples of model performance decreasing with increased dissimilarity between training and test data, for co-folding (left) and binding affinity prediction (right).
Molecule Networks: data visualization using PyVis
Over the past few years I have explored different data visualization strategies with the goal of rapidly communicating information to medicinal chemists. I have recently fallen in love with “molecule networks” as an intuitive and interactive data visualization strategy. This blog gives a brief tutorial on how to start generating your own molecule networks.
A tougher molecular data split – spectral split
Scaffold splits have been widely used in molecular machine learning which involves identifying chemical scaffolds in the data set and ensuring scaffolds present in the train and test sets do not overlap. However, two very similar molecules can have differing scaffolds. In an example provided by Pat Walters in his article on splitting chemical data last month, he provides an example where two molecules just differ by a single atom and thus have a very high Tanimoto similarity score of 0.66. However, they have different scaffolds (figure below).
In this case, if one of the molecules were in the train set and the other in the test set, predicting the test molecule would be quite trivial as there is data leakage. Therefore, we need a better splitting method such that there is minimal overlap between the train and test set. In this blogpost, I will be discussing spectral split, a splitting method introduced by our fellow OPIG member, Klarner et. al (2023).
Spectral split
Spectral split or clustering is based on the spectral graph partitioning algorithm. The basic idea of spectral clustering is as follows: The dataset is projected on a R^n matrix. An affinity matrix using a kernel that could be domain-specific is defined. Following that, the graph Laplacian is computed from the affinity matrix, followed by its eigendecomposition. Then, k eigenvectors corresponding to the k lowest/highest eigenvalues are selected. Finally, the clusters are formed using k-means.
In the context of molecular data splitting, one could use the Tanimoto similarity metric to construct a similarity matrix between all the molecules in the dataset. Then, a spectral clustering method could be used to partition the similarity matrix such that the similarity within the cluster is maximized whereas the similarity between the clusters is minimized. Spectral split showed the least overlap between train (blue) and test (red) set molecules compared to scaffold splits (figure from Klarner at. al. (2024) below)
In addition to spectral splits, one could attempt other tougher splits one could attempt such as UMAP splits suggested by Guo et. al. (2024). For a detailed comparison between UMAP splits and other commonly used splits please refer to Pat Walters’ article on splitting chemical data.
The XChem trove of protein–small-molecules structures not in the PDB
The XChem facility at Diamond Light Source is truly impressive feat of automation in fragment-based drug discovery, where visitors comes clutching a styrofoam ice box teeming with apo-form protein crystals, which the shifter soaks with compounds from one or more fragment libraries and a robot at the i04-1 beamline kindly processes each of the thousands of crystal-laden pins, while the visitor enjoys the excellent food in the Diamond canteen (R22). I would especially recommend the jambalaya. Following data collection, the magic of data processing happens: the PanDDA method is used to find partial occupancy in the density, which is processed semi-automatedly and most open targets are uploaded in the Fragalysis web app allowing the ligand binding to be studied and further compounds elaborated. This collection of targets bound to hundreds of small molecules is a true treasure trove of data as many have yet to be deposited in the PDB, making it a perfect test set for algorithm design: fragments are notorious fickle to model and deep learning models cannot cheat by remembering these from the protein database.
Continue readingOut of the box RDKit-valid is an imperfect metric: a review of the KekulizeException and nitrogen protonation to correct this
In deep learning based compound generation models the metric of fraction of RDKit-valid compounds is ubiquitous, but is problematic from the cheminformatics viewpoint as a large fraction may be driven by pyrrolic nitrogens (see below) rather than Texas carbons (carbon with 5 bonds like the Star of Texas). In RDKit, no error is more irksome that the KekulizeException or ValenceException from RDKit sanitisation. These are raised when the molecule is not correct. This would make the RDKit-valid a good metric, except for a small detail: the validity is as interpreted from the the stated implicit and explicit hydrogens and formal charges on the atoms, which most models do not assign. Therefore, a compound may not be RDKit-valid because it is actually impossible, like a Texas carbon, but in many cases it is because the formal charge or implicit hydrogen numbers of some atoms are incorrect. In both case, the major culprit is nitrogen. Herein I go through what they are and how to fix them, with a focus on aromatic nitrogens.
Tanimoto similarity of ECFPs with RDKit: Common pitfalls
A common measure for the similarity of two molecules is the Tanimoto similarity of their ECFPs (Extended Connectivity FingerPrint). However, there is no clear standard in literature for what kind of ECFPs should be used when calculating the Tanimoto similarity, and that choice can lead to substantially different results. In this post I wish to shed light on some results you should know about before you jump into your calculations.
A blog post on how ECFPs are generated was written by Marcus Dablander in 2022 so please take a look at that. In short, ECFPs have a hyperparameter called the radius r, and sometimes a fingerprint length L. Each entry in the fingerprint indicates the presence or absence of a particular substructure in the molecule of interest, and the radius r defines how large the substructures that you consider are. If you have r=3 then you consider substructures made by going up to three hops away from each atom in your molecule. This is best explained by this figure from Marcus’ post:
I really hope my compounds get the green light
As a cheminformatician in a drug discovery campaign or an algorithm developer making the perfect Figure 1, when one generates a list of compounds for a given target there is a deep desire that the compounds are well received by the reviewer, be it a med chemist on the team or a peer reviewer. This is despite scientific rigour and training and is due to the time invested. So to avoid the slightest shadow of med chem grey zone, here is a hopefully handy filter against common medchem grey-zone groups.
Continue readingSort and Slice Tutorial – An alternative to extended connectivity fingerprints
Background¶
Sort and Slice (SNS) was developed by a former OPIGlet, Markus, as a method for improving Extended Connectivity Fingerprints (ECFPs) by overcoming bit collisions. ECFPs are a form of topological fingerprint which denote the absence and presence of circular substructures in a molecule. The steps for deriving an ECFP from a molecule are as follows:
Identifier assignment:
Each atom in the molecule is assigned an initial numerical identifier; this is typically generated by hashing a tuple of atomic properties called Daylight atomic invariants into a 32-bit integer. These properties are:
- Number of non-hydrogen neighbours.
- Valence – number of neighbouring hydrogens.
- Atomic number.
- Atomic mass.
- Atomic charge.
- Number of hydrogen neighbours.
- Ring membership.*
*Ring membership is an additional property that is often used but is not one of the original Daylight atomic invariants.
Continue reading