class: center, middle, inverse, title-slide # SSI fellowship application ## Link to slides: <a href="http://bit.ly/ssi-app" class="uri">http://bit.ly/ssi-app</a> ### Sam Abbott (<span class="citation">@seabbs</span>) ### London School of Hygiene and Tropical Medicine --- class: inverse # Outline ## 1. Who am I? ## 2. Why am I applying for an SSI fellowship? ## 3. What would I do with an SSI fellowship --- class: center, middle, inverse # Who am I? --- # Background * Research Fellow in real-time modelling of infectious disease outbreaks at the London School of Hygiene and Tropical Medicine. * PhD in infectious disease modelling focussing on the optimal usage of the BCG Tuberculosis vaccine. * Working primarily on Covid-19 forecasting, nowcasting, and reactive real-time analysis. * Strong interest in open science and science as software. --- # Current focus - 3 main categories 1. **Developing methodology** to estimate the [basic](https://wellcomeopenresearch.org/articles/5-17/v1) and [effective](https://wellcomeopenresearch.org/articles/5-112/v2) reproduction (Rt) of Covid-19 and provide short term forecasts of key indicators (cases, hospital admissions, deaths, hospital bed usage etc.). 2. **Deploying this methodology** as [software](https://epiforecasts.io/EpiNow2/) for use by others and by [ourselves in collaboration with the Met office](https://epiforecasts.io/covid/)¹. Software used by the WHO, ECDC, and [several groups supporting other governments](https://elifesciences.org/articles/58785). Nowcasts/Forecasts used by the CDC, WHO, UK Government, ECDC etc. 3. **[Real-time analysis](https://scholar.google.co.uk/citations?hl=en&user=GqZm90IAAAAJ&view_op=list_works&sortby=pubdate) used by policy makers (SPI-M, SAGE, NERVTAG):** ² * Weekly estimates of the effective reproduction number, current infections and forecasts/projections of future admissions etc. * Developments in the epidemic, for example tracking nosocomial transmission in April 2020. * The increase in transmissibility and severity measures of the [B.1.1.7 variant](https://github.com/epiforecasts/covid19.sgene.utla.rt). .footnote[ [1] [https://epiforecasts.io/covid/](https://epiforecasts.io/covid/) [2] [http://bit.ly/sea-scholar](http://bit.ly/sea-scholar) ] --- class: center, middle, inverse Why am I applying for an SSI fellowship? --- # Why am I applying for an SSI fellowship? * The Covid-19 pandemic has highlighted the need for combining robust software engineering with epidemological knowledge. * Many of the real time analysis projects from the beginning of the pandemic have been shuttered, likely due to technical debt. * Most of the forecasts that are used by policy makers come from models that are closed source or not fully documented/supported making them only open source in theory. * Many researchers lack the skills required to make their methods open source and there is little incentive to do so. * Research software engineering when considered at all is thought about as packaging of already complete research work rather than a truly integrated part of the process. --- class: center, middle, inverse In trying to respond to the Covid-19 pandemic I have realised that I simply do not have the software engineering skills required to do so effectively and that this is a systemic problem in infectious disease modelling that hampers our ability to provide advice to policy makers. --- class: center, middle, inverse # What would I do with the fellowship? --- # Personal * Use the SSI network to enhance my own software engineering skills so I can more effectively do my work. * In particular, I would like to expand beyond the silo of R knowledge into the wider community of practice. Infectious disease modelling is generally R centric and I feel this means that innovations from other fields are missed. * Use some of the fellowship funding for skills training (ideally a Python software development course) and potentially funds for conference travel (such as the Stan conference). --- # Community * Currently in the process of setting up a software engineering discussion group within LSHTM. Fellowship funding would be a great way to provide customised training to this group many of whom are transitioning from more theoretical areas of research. * Interested in potentially highlighting some of the key lessons learnt from the Covid-19 pandemic on the importance of software engineering in research using a short professional animation combined with a blog post. Fellowship funding could be used for this. * Advocate for researchers to consider their work as software rather than as a step towards a peer reviewed paper. * Strongly advocate for RSEs to be hired within LSHTM and in the wider infectious disease modelling community. --- class: inverse # Summary * Researching real time analysis methods for infectious diseases. Focussing on the Covid-19 pandemic. * Applying as the response from the infectious disease modelling community to the pandemic has shown the impact of not having software engineering skills available. * I would use the fellowship to enhance my own knowledge, to advocate for RSEs in infectious disease modelling, support developing a community of practice at LSHTM, and potentially to summarise some of the lessons on the importance of research engineering learnt from the Covid-19 pandemic using a professional animation.