Three weeks ago I set off to Basel for a summer school at Roches' Predictive Modeling and Data Analytics chapter. Even though I was really looking forward to this time working on something different from my PhD project, I was not expecting the week to leave me with so many positive impressions.

🧠 Most of all, I loved working on the challenging project of protein turnover prediction with the many talented people from the field of computational biology!

📈 Seeing how much progress we made in just a few days, and how our skills were so distinct but still came together to create a well-rounded result really impressed me. It reminded me how fun and productive collaboration can be in research, or to put it in our team leads' words (originally a quote from Phil Jackson about team spirit in Basketball, but it applies here as much as in sports):

The strength of the team is each individual member. The strength of each member is the team.

🏆 This really proved true for us, considering that we would have never managed to produce such an amount of results on our own, and how well our expertises complemented each other in this project. I am very thankful for this experience.

I  presented my recent work on antibiotic susceptibility prediction at the European Congress of Clinical Microbiology and Infectious Diseases.

🧬 Whole genome sequencing (WGS) is increasingly being adapted for rapid susceptibility testing since it works on a much shorter time scale than phenotypic testing (a few hours vs days -  weeks) and it outperforms other genetics based approaches by considering the entire genome.

🧠 My work on graph convolutional networks for resistance prediction in E. coli will complement existing WGS prediction approaches by being able to predict the effect of unknown mutations.

📋 It will thereby be an addition to the catalogue-based genetic approach of WGS for resistance prediction, which considers only known resistance alleles and mutations.

This was my first international conference and I was fortunate to get to see what the infectious disease field looks like outside of the UK.

I gave a flashtalk at the POP UP conference of MicrobesNG in Oxford. These are microbial genomics conferences aimed at early career researchers that take place in different cities all of the UK. 

🧠 I talked about our new project, working on graph convolutional networks for antibiotic susceptibility prediction. By including the protein structure of the targets of specific antibiotics in the prediction process, we can implicitly include information about the structural changes that follow mutations. 

🧫 Currently, we are working with datasets from M. tuberculosis and E. coli, but we are hoping to apply this technique to many other clinically relevant pathogens in the future.

🩺   In the long run, these methods can complement the slow phenotypic and catalogue-based genetic approaches to susceptibility prediction, thereby closing some gaps in the diagnostics process.

It was a pleasure to write this short blog post about my latest paper and the paper-writing process for the microbiology society.

🖊️ Apart from a very short summary of the key findings of the paper I hence also describe parts of my PhD journey and what led my supervisor and I towards this particular project.

📎 In case you are interested in the full story on fitness compensation in TB, here's the link to the paper: https://doi.org/10.1099/mgen.0.001187

Our recent paper on fitness compensation in M. tuberculosis (Mtb) has found its new home at Microbial Genomics. I am really excited to see this work published in a journal of the Microbiology Society!

https://doi.org/10.1099/mgen.0.001187

This is my first first-authored paper, as well as the first paper published during my PhD and I am really happy with how the project turned out.

Key take-aways:

📋 In this work, we construct a comprehensive list of mutations that are significantly associated with resistance, based on a collection of > 70,000 clinical Mtb samples.

📉 On top of that, we show that in vitro growth data can recapitulate the fitness cost of antibiotic resistance...

📈 ... and also the restoring effect of compensatory mutations in some Mtb lineages.

📍 Lastly, we investigate the clustering of compensatory mutations in specific regions of the antibiotics target protein, the RNA polymerase.

Our paper investigating compensatory mutations in M. tuberculosis and their influence on in vitro fitness has just been accepted at Microbial Genomics, a journal published by the Microbiology Society.

🎉  We are really happy that this work has found a new home and hope to see it out and published any time soon. 

Our work on compensatory mutations (CMs) in M. tuberculosis has just been uploaded as a preprint. The paper describes the large-scale identification of CMs based on a collection of > 70,000 clinical TB samples and identifies CM clustering regions on the RNA polymerase. 

🧫 We also attempt to dissect the in vitro growth densities associated with the presence of resistance mutations and CMs. This can be used as a proxy for the previously described fitness cost and fitness salvage associated with resistance mutations and CMs, respectively. 

📈 We found a strong association of CM presence with higher than wild type in vitro growth densities for the more virulent Lineages 2 and 3, but the underlying cause is not straightforward to identify. Further work will be needed to disentangle the influence of lineage- and clade-association on the observed growth levels.

The OxBacNet meetings brings together bacteriologists from all of Oxfords research institutions.  Every term, some senior and junior scientists have the possibility to present their work to members of the network at the OxBacNet symposium. 

🧬 I was happy to present the final stages of my work on compensatory mutations in M. tuberculosis at this Trinity term edition.

🖨️  This work is now almost concluded and will soon be published as a preprint, with all data and conducted analysis available online on my GitHub page.

I was able to showcase my work on compensatory mutations in the RNA polymerase in M. tuberculosis at this years Multidisciplinary approaches to AMR conference in Oxford. 

🦠 My preliminary results indicate an interesting association between the presence of compensatory mutations in TB samples and unusually high sample growth in vitro. These findings could very well explain a part of the observed increased transmissibility of Lineage 2 M. tuberculosis in high burden countries and should be monitored further.

🥇 As this was my very first conference poster, I was very happy to be awarded a poster price for my contribution.