Viral Escape

The moment is almost here. By now, nearly everyone in the Netherlands has had the chance to get vaccinated, the 1.5-metre rule is being lifted, and the government aims to lift the other measures too. That sounds like the end of COVID-19, doesn’t it? Unfortunately, SARS-CoV-2 (the coronavirus that causes COVID-19) mutates rapidly, much like the flu virus. Among those thousands of mutations, we have already identified several variants of SARS-CoV-2, including the alpha, beta, gamma and (currently the most common) delta variant. Fortunately, most vaccines are also effective against the new variants of the virus, but what if that changes? A "viral escape" is a worst-case scenario in which the virus mutates just enough that existing antibodies no longer recognise it. The consequences are serious: such a dangerous mutation would evade the immune systems of people who have been vaccinated (or previously infected). In short, we would be back to square one.

How AI can help

To avoid this scenario, it is important to identify which of the thousands of mutations could actually pose a serious threat, so that vaccine development can respond as quickly as possible. AI can help with this. MIT researchers have devised a new way to model ‘viral escapes’ using models originally developed to analyse language: the well-known Natural Language Processing (NLP) models.

The idea behind using an NLP model in this case is as follows: viruses mutate in ways that follow the biological rules of protein structure while also being as advantageous as possible. SARS-CoV-2, for example, mutates in the spike protein shown in the image below. People who have recently had COVID-19 or been vaccinated currently have antibodies that fit the spike protein. As a result, SARS-CoV-2 virus cells can no longer make contact with human cells, and someone becomes infected. The virus therefore aims to mutate its spike proteins again quickly, so that the antibodies no longer fit but the receptors on human cells still do. It needs to mutate in a way that lets it escape the human immune system (the antibodies) without dying or losing its ability to reproduce. Similarly, for an NLP model, a sentence must not only have the right meaning (semantics), but also be grammatically correct (syntax). Using those same two principles, the researchers have creatively adapted NLP models to detect changes in the genetic code of viruses.

An example

The example below illustrates how the NLP model assesses which virus mutations could lead to viral escape. The first sentence represents the virus before it mutates. The second sentence (from the left) shows a small mutation. Its meaning has barely changed, and it is grammatically correct. With this mutation, the new virus still resembles the original closely enough for the immune system to recognise and attack it. No new antibodies are needed in this case. The third sentence is grammatically incorrect. In the language of the virus, such a mutation would be considered unsuccessful. The final sentence is where the danger lies. It is grammatically correct and has the right semantics. These are the exceptions that the NLP model predicts could lead to viral escape. The researchers call the search for these exceptions 'constrained semantic change search' (CSCS).

NLP model and training

Of course, the actual NLP model was not trained on sentences, but on the building blocks of spike proteins from different coronaviruses: amino acid sequences. The training set contained just under 1,000 sequences of the SARS-CoV-2 spike protein and another 3,000 spike protein amino acid sequences from other types of coronavirus. The NLP model is shown in the image below. Internally, the model constructs a semantic representation, or “embedding”, for a given amino acid sequence. The model’s output shows how well an amino acid fits the “grammar” of the sequence. In the image, the amino acid that fits the sequence best grammatically is marked with a capital A.

Testing

Of the 891 different coronavirus spike protein amino acid sequences the researchers examined with the model, one came from a strain that had reinfected someone who had recovered from Covid-19 the previous year. This sequence therefore quickly received a high CSCS score. Only three other sequences in the set showed both greater semantic change and higher so-called grammaticality. The researchers also fed some of the new variants into their algorithm and found that both the South African and British strains scored "fairly high" for their likelihood of escape.

How viral escape can be prevented

What now? If new mutations are closely monitored and the algorithm is used to assess how dangerous they are, researchers can test suspect strains in the laboratory as quickly as possible and adapt vaccines accordingly. The testing works like this: suspect strains and antibodies are brought together. If the antibodies do not bind to the spike proteins, the current antibodies no longer provide protection. It is still unclear how much time vaccine developers actually save with an AI-based approach like this. What we do know is that in a pandemic as large as this one, every second counts.

References

1. Hie, Brian, et al. "Learning the language of viral evolution and escape." Science371.6526 (2021): 284-288. DOI: 10.1126/science.abd7331
2. https://singularityhub.com/2021/01/19/a-language-ai-is-accurately-predicting-covid-19-escape-mutations/
3. https://spectrum.ieee.org/ai-predicts-most-potent-covid-19-mutations
4. https://news.mit.edu/2021/model-viruses-escape-immune-0114
5. https://qz.com/africa/1995639/scientists-use-algorithms-to-predict-new-covid-19-variants/