Thursday, April 24, 2014

Predicting the next Flu Season

It happens every year; either you or someone you know gets the flu. And then it spreads. Suddenly every second person seems to be feeling sick, having passed their illness on to those around them. Wouldn’t it be great if we could predict which virus was causing the problem and tailor the flu vaccine to preemptively protect you? Up until recently this sort of prediction was done by humans who made logical guesses as to which strain of the virus was most likely to rear its ugly head.


A new paper by two computational biologists from the University of Columbia explains how they hope to change that. Their approach was to outline all the different families of the flu virus and create predictive models of their potency and compare their genetic drift. Using this model they were able to predict with 93 percent accuracy which families will be the most widespread in the upcoming year. As the article in Popular Mechanics explains this is not a cure for the flu, nor does it solve problems of unknown mutations appearing in variants of the virus, however it does help to improve the vaccine and increase its effectiveness.


While mutations in flu viruses certainly pose a problem, knowing which family to target should be more effective as similar strains are still vulnerable to the same vaccine. The biggest challenge with new mutations is when they affect the shell proteins of the virus, allowing them to hide within an organism that doesn’t recognize their markings. This too is taken into account by the model as certain families of viruses are more likely to mutate in this way compared to others.


Nonetheless there is still a long way to go with the model. It demonstrates a strong proof of concept, especially considering how volatile the real world can be, but it still fell short in a number of areas. Particularly in estimating the seasonal decline of a flu family, where it was only correct three out of four times. This new model for predicting the strength and likelihood of a particular virus is an exciting new tool for scientists as they continue to fight against seasonal outbreaks.

Posted by: Kirk MacKinnon (B)

4 comments:

  1. The tracking of the flu may not help to stop the virus from evolving, but maybe if this model can track to the exact place where the virus evolved, they could understand why it happened. From this, they could aim to study the mechanisms that have mutated and evolved and how populations of viruses became resistant to the past flu shot. Maybe there are environmental factors or human interaction that create conditions for this evolution to be favored. I think this can go a lot farther than just spreading awareness.

    -Nicole B

    ReplyDelete
    Replies
    1. Its exciting to think that this might lower the incidence of a potentially fatal illness. Being able to predict what strains of the flu virus will be the most potent combined with the flu vaccine could very much alter how we treat it in the future.

      Posted by: Kirk M

      Delete
  2. This is an exciting study. Although I do not get the flu shot every year, I know many people who do. With this new study, maybe there will be even further ways to prevent a person from getting the flu. Being able to track where the flu originated from will hopefully drop the percentage of people who get flu a year.

    -Amber Vien

    ReplyDelete
    Replies
    1. If we can accurately predict which strain will pose the most problem it would infinitely improve how effective the flu vaccine is. Instead of treating the flu as a natural part of life we could easily and cheaply avoid getting it by simply getting a yearly flu vaccine. This is a very exciting step forward.

      Posted by: Kirk M

      Delete