The q-distance

As we begin to recover from the COVID-19 pandemic, a key question is if we can avert such disasters in future. Unfortunately, current surveillance protocols are limited to qualitative impact assessments of viral diversity and habitat encroachments. These efforts, while vital for ecosystem monitoring, do not aim to precisely quantify emergence risk across hosts and viral families. In the state-of-the-art, closeness of strains is measured by their edit distance, e.g. how many mutations makes an avian flu strain human-adapted. However, this ignores the actual odds of those mutations in the wild, and is blind to the true jump risk. Thus, sampling viral diversity in the wild currently gives us no indication of which strains are more risky or when we might be at the edge of emergence. In this study, we develop a more meaningful metric for sequence comparison. Our metric, the q-distance, precisely quantifies the probability of spontaneous jump by random chance. Learning from patterns of mutations from large sequence databases, the q-distance adapts to the specific organism, the background population, and realistic selection pressures. As our first application, we demonstrate that the q-distance enables improved forecasting of future strains in the context of the seasonal Influenza epidemic in both hemispheres over World Health Organization (WHO) recommendations for the flu-shot composition nearly consistently over past two decades. As our second application, we investigate SARS2 origin, and precisely quantify the likelihood of different animal species that hosted an immediate progenitor. In contrast to a qualitative narrative, we produce a list of related species of bats that have a quantifiably high likelihood of being the source. Additionally, we identify specific rodents with a credible likelihood of hosting a SARS2 ancestor. Combining machine learning and large deviation theory, the analysis framework reported here can open the door to actionable predictions of future pandemics.

Seasonal Epidemic Influenza

SARS2 Origin Story

To Do

Improve Dominant Strain Tracking

- Improve Northern Hemisphere Influenza predictions

Fast Notebook Example

- Make a python package that computes distance between current (2019/20) Influenza HA and NA strains
- Make a python package for COVID spike sequences
- We need to make sure that we have a membership test in this packages
- We package the qnet data so that no qnet computation is taking place
- Can we write this part in C/C++ (the distanec calcualtion part)?