Developments in COVID19: Part II
COVID-19 unites efforts in research
Since the early days of the virus spread throughout the world, research at the international level also began. Proof of this is the different consortiums created to fight the pandemic, such as COVID-19 High-Performance Computing (HPC) Consortium, COVID-19 Genomics (COG-UK) Consortium or Host Genetics Initiative (HGI). The latter, in which 24Genetics participates through a collaboration with the Health Research Institute of the La Paz University Hospital (IdIPAZ), groups together almost 200 international research projects focused on genetic factors of susceptibility to the virus. This new entry will describe the history of genetic discoveries and the development of the research that 24Genetics is currently carrying out.
24Genetics' collaborative research contribution
Thanks to the consortium in which 24Genetics and (HGI) participate, we discovered many and varied genetic markers. As new data is available, the genetic results have become more meaningful. One of the first studies that saw the light, with the help of HGI, identified critical markers in the antiviral defence and the inflammatory processes. These included specific chromosomes and genes such as FYCO1 (Chr. 3), IGF1 (Chr. 12) or DPP9 (Chr. 19) (Figure 1).
Figure 1. Pairo-Castineira, E., et al. Genetic mechanisms of critical illness in Covid-19. medRxiv 2020.09.24.20200048; doi: https://doi.org/10.1101/2020.09.24.20200048
Results of the research with IdiPaz
No less significantly, and with a much larger number of samples, the importance of blood group as an immunoprotective factor was raised (2). The ABO gene (Chr.3), the main actor, came out showing potential protection of between 14% and 16% in particular blood groups. The research also brought to the table more inflammatory genetic factors related to respiratory failure (Figure 2).
Figure 2. The Severe Covid-19 GWAS Group. Genomewide Association Study of Severe Covid-19 with Respiratory Failure. 2020 N Engl J Med 2020; 383:1522-1534 DOI: 10.1056/NEJMoa2020283
And what does 24Genetics contribute to all this?
In October, 24Genetics takes the baton, and with the fieldwork done by IdIPAZ and the samples already sequenced by our reference laboratories, it gets to work. One of the first tasks carried out was the filtration and "healing" of the data generated. One of the processes started was the analysis of genetic data from people with IDOC disease concerning people in intensive care (Figure 3).
Figura 3. Figura proporcionada por Lasse Folkersen, BS y PhD
It was then decided to begin using artificial intelligence by analyzing the set of clinical variables reported, to know closely, which ones were more or less important (Table 1).
Table 1. Chart provided by Carlos Ortega, Data Scientist
Artificial intelligence, on this occasion, made it possible to classify the variants by significance. In the following table, you can see the different factors obtained. However, it was necessary to validate the model; for that, we used SHAP (Shapley Additive ExPlanations). In short, this model allows the algorithm to explain the weight that each factor has (Table 2).
Table 2. Chart provided by Carlos Ortega, Data Scientist
Neutrophil ratio, blood saturation or interleukin-6 (parameters that have already been put under the microscope before) were significant parameters confirmed (3,4,5). Once we had the data, the next step was to train the algorithm to understand the potential predictive value.
Benefits of the application of the algorithm for patients with COVID-19
Finally, we trained the algorithm to predict, with a 97% coincidence, the mortality or survival of COVID patients. Despite the encouraging preliminary results, the scientific team of 24Genetics is cautious and does not lose sight of possible limiting biases. Also, 24Genetics will validate in datasets with several necessary samples to verify the data obtained. If all the findings are validated, the tool could be useful in the survival of COVID patients or the improvement of hospital protocols. We performed all the computing process with open source code. Therefore to be used without cost in any clinic and hospital in the world. In the meantime, the team will keep on researching.
- Instituto de Investigación Sanitaria del Hospital Universitario La Paz (IdIPAZ)
- Carlos Ortega, Data Scientist
- Lasse Folkersen, BS and PhD
- The COVID-19 Host Genetics Initiative. The COVID-19 Host Genetics Initiative, a global initiative to elucidate the role of host genetic factors in susceptibility and severity of the SARS-CoV-2 virus pandemic. J. Hum. Genet. 28, 715–718 (2020). https://doi.org/10.1038/s41431-020-0636-6.
- Pairo-Castineira, E., et al. Genetic mechanisms of critical illness in Covid-19. medRxiv 2020.09.24.20200048; doi: https://doi.org/10.1101/2020.09.24.20200048
- The Severe Covid-19 GWAS Group. Genomewide Association Study of Severe Covid-19 with Respiratory Failure. 2020 N Engl J Med 2020; 383:1522-1534 DOI: 10.1056/NEJMoa2020283
- Protasio Veras, F., et al. SARS-CoV-2–triggered neutrophil extracellular traps mediate COVID-19 pathology. J Exp Med 7 December 2020; 217 (12): e20201129. doi: https://doi.org/10.1084/jem.20201129
- Shenoy, N., Luchtel, R. & Gulani, P. Considerations for target oxygen saturation in COVID-19 patients: are we under-shooting?. BMC Med 18, 260 (2020). https://doi.org/10.1186/s12916-020-01735-2
- Chen X, et al. Detectable Serum Severe Acute Respiratory Syndrome Coronavirus 2 Viral Load (RNAemia) Is Closely Correlated With Drastically Elevated Interleukin 6 Level in Critically Ill Patients With Coronavirus Disease 2019. Clin Infect Dis. 2020 Nov 5;71(8):1937-1942. doi: 10.1093/cid/ciaa449. PMID: 32301997; PMCID: PMC7184354.