Tag Archives: Graphs

Graphormer: Merging GNNs and Transformers for Cheminformatics

This is my first OPIG blog! I’m going to start with a summary of the Graphormer, a Graph Neural Network (GNN) that borrows concepts from Transformers to boost performance on graph tasks. This post is largely based on the NeurIPS paper Do Transformers Really Perform Bad for Graph Representation? by Ying et. al., which introduces the Graphormer, and which we read for our last deep learning journal club. The project has now been integrated as a Microsoft Research project.

I’ll start with a cheap and cheerful summary of Transformers and GNNs before diving into the changes in the Graphormer. Enjoy!

Continue reading →

Universal graph pooling for GNNs

Graph neural networks (GNNs) have quickly become one of the most important tools in computational chemistry and molecular machine learning. GNNs are a type of deep learning architecture designed for the adaptive extraction of vectorial features directly from graph-shaped input data, such as low-level molecular graphs. The feature-extraction mechanism of most modern GNNs can be decomposed into two phases:

  • Message-passing: In this phase the node feature vectors of the graph are iteratively updated following a trainable local neighbourhood-aggregation scheme often referred to as message-passing. Each iteration delivers a set of updated node feature vectors which is then imagined to form a new “layer” on top of all the previous sets of node feature vectors.
  • Global graph pooling: After a sufficient number of layers has been computed, the updated node feature vectors are used to generate a single vectorial representation of the entire graph. This step is known as global graph readout or global graph pooling. Usually only the top layer (i.e. the final set of updated node feature vectors) is used for global graph pooling, but variations of this are possible that involve all computed graph layers and even the set of initial node feature vectors. Commonly employed global graph pooling strategies include taking the sum or the average of the node features in the top graph layer.

While a lot of research attention has been focused on designing novel and more powerful message-passing schemes for GNNs, the global graph pooling step has often been treated with relative neglect. As mentioned in my previous post on the issues of GNNs, I believe this to be problematic. Naive global pooling methods (such as simply summing up all final node feature vectors) can potentially form dangerous information bottlenecks within the neural graph learning pipeline. In the worst case, such information bottlenecks pose the risk of largely cancelling out the information signal delivered by the message-passing step, no matter how sophisticated the message-passing scheme.

Continue reading →

Exploring topological fingerprints in RDKit

Finding a way to express the similarity of irregular and discrete molecular graphs to enable quantitative algorithmic reasoning in chemical space is a fundamental problem in data-driven small molecule drug discovery.

Virtually all algorithms that are widely and successfully used in this setting boil down to extracting and comparing (multi-)sets of subgraphs, differing only in the space of substructures they consider and the extent to which they are able to adapt to specific downstream applications.

A large body of recent work has explored approaches centred around graph neural networks (GNNs), which can often maximise both of these considerations. However, the subgraph-derived embeddings learned by these algorithms may not always perform well beyond the specific datasets they are trained on and for many generic or resource-constrained applications more traditional “non-parametric” topological fingerprints may still be a viable and often preferable choice .

This blog post gives an overview of the topological fingerprint algorithms implemented in RDKit. In general, they count the occurrences of a certain family of subgraphs in a given molecule and then represent this set/multiset as a bit/count vector, which can be compared to other fingerprints with the Jaccard/Dice similarity metric or further processed by other algorithms.

Continue reading →

How to turn a SMILES string into a molecular graph for Pytorch Geometric

Despite some of their technical issues, graph neural networks (GNNs) are quickly being adopted as one of the state-of-the-art methods for molecular property prediction. The differentiable extraction of molecular features from low-level molecular graphs has become a viable (although not always superior) alternative to classical molecular representation techniques such as Morgan fingerprints and molecular descriptor vectors.

But molecular data usually comes in the sequential form of labeled SMILES strings. It is not obvious for beginners how to optimally transform a SMILES string into a structured molecular graph object that can be used as an input for a GNN. In this post, we show how to convert a SMILES string into a molecular graph object which can subsequently be used for graph-based machine learning. We do so within the framework of Pytorch Geometric which currently is one of the best and most commonly used Python-based GNN-libraries.

We divide our task into three high-level steps:

  1. We define a function that maps an RDKit atom object to a suitable atom feature vector.
  2. We define a function that maps an RDKit bond object to a suitable bond feature vector.
  3. We define a function that takes as its input a list of SMILES strings and associated labels and then uses the functions from 1.) and 2.) to create a list of labeled Pytorch Geometric graph objects as its output.
Continue reading →

Issues with graph neural networks: the cracks are where the light shines through

Deep convolutional neural networks have lead to astonishing breakthroughs in the area of computer vision in recent years. The reason for the extraordinary performance of convolutional architectures in the image domain is their strong ability to extract informative high-level features from visual data. For prediction tasks on images, this has lead to superhuman performance in a variety of applications and to an almost universal shift from classical feature engineering to differentiable feature learning.

Unfortunately, the picture is not quite as rosy yet in the area of molecular machine learning. Feature learning techniques which operate directly on raw molecular graphs without intermediate feature-engineering steps have only emerged in the last few years in the form of graph neural networks (GNNs). GNNs, however, still have not managed to definitively outcompete and replace more classical non-differentiable molecular representation methods such as extended-connectivity fingerprints (ECFPs). There is an increasing awareness in the computational chemistry community that GNNs have not quite lived up to the initial hype and still suffer from a number of technical limitations.

Continue reading →

Calculating symmeterised small molecule RMSDs using graph automorphisms in python with GEMMI and NetworkX

When a ring flips, how do we calculate RMSD?

This surprisingly simple question leads to a very interesting problem! If we take a benzene molecule, say, and rotate it 180 degrees, then we have the exact same molecule, but if we have a data structure in which our atoms are labelled, and we apply the same transformation to the atomic positions, the numbering does not reflect that symmetry. If we were then naively to calculate the RMSD it would be huge, despite the fact that the molecule is, chemically speaking, identical.

How can we make our RMSD calculations reflect these symmetries?

Continue reading →