Bioinformatics: A way to deciphere DNA and cure life's deadliest diseases | Spencer Hall | TEDxUGA

in 1799 while constructing a Ford in Egypt a group of French soldiers came across an ancient Egyptian tablet containing hieroglyphics on it what made the discovery this tablet so groundbreaking was not just the fact that it contained hieroglyphics because by this point quite a few stones were that script had already been discovered it was what came with the hieroglyphics that made this stone so important Greek because while for over a thousand years no one had been able to read the ancient Egyptian script everyone trained in classical languages could read Greek this discovery 40 years later enabled the French scholar jean-francois Champollion to finally decode the meaning of the hieroglyphics the symbols which had been silent for over a millennium became a new window into ancient Egyptian life and culture today in the first few decades of the 21st century we are faced with a challenge and an opportunity similar to that of the rosetta stone it comes from a language which Dwarfs any human language in the immense number of volumes written in it and that's the language of DNA DNA contains the blueprints for every biological process of every living thing on earth and even a small defect in the genetic code can cause devastating health consequences understanding how genes work both individually and with each other is crucial to the development of 21st century personalized medicine the problem is that even when we actually have an organism's DNA in front of us it's often not easy to tell what an individual gene does even more difficult can be teasing out the subtle influences that multiple genes have on each other but answering those kinds of questions is foundational to attacking genetically influenced diseases at the root cause forty years ago even if we had had all the genetic information that we have today which by the way we didn't most of its been found in the last 20 years but even if we had we could not have even begun to sift through it an army of people would never have had time to sift through the enormous amount of genetic data needed to answer health-related questions today however a new subfield of biology is being developed that is making what formerly seemed impossible into a reality that field is bioinformatics bioinformatics is the application of statistics and information theory to genetic data it allows us to sift through an enormous amount of that data in a way that would be impossible to do manually let me give you an example of how bioinformatics can be useful there's a very common classic statistical procedure called the two-sample t-tests and its purpose is if we have two groups and we want to know if the average number of individuals with a certain feature in this group is the same as the average number of individuals with that feature in this group we run the test to find out so for example if I have two jars of marbles red marbles and blue marbles in each jar and I want to know if the average number of red marbles and a handful from this jar is going to be the same as the average number of red marbles and a handful from this jar I can run the test by taking a sample of maybe 30 marbles from this charm 30 marbles from this jar and then counting up the number of times red marbles appear in each handful once I have those two numbers I plug them back into the ugly equation you just saw and what I get back is the probability that I have this number of red marbles and dis hand and this number of red marbles and this hand that we just got if both jars have the same number of red marbles let me seen arcane but let me give you an example of how we can use it in bioinformatics suppose we want to find out whether a particular gene is associated with thyroid cancer we could take its sample of maybe the genomes of a hundred people with the cancer and a hundred healthy people and then count the number of times a gene appears in each group once we have those two numbers just like with the marbles we plug them back into the equation and what we get back remember is the probability that data that gene appears this many times in the cancer group and this many times in the healthy group if you're equally likely to have the gene whether you have the cancer or not and if that probability is sufficiently low we usually use the cutoff point of maybe 5% the we can conclude that it's very likely that gene does have some sort of association with the cancer this is a paradigm example of what makes bioinformatics so useful we started with an ocean of genetic data the four genomes of 200 people and using statistics have narrowed our focus down to just one gene that can then be further studied by geneticists now I've given you ooh a little bit of an idea of how we can use bioinformatics but you may still be wondering why a specifically genetic understanding of how disease works is important and answer that question I want to take one particularly cruel condition as an example cystic fibrosis is a genetically inherited mutation that causes progressive lung disease and the way that works is everybody's lungs CF or not have a thin mucous lining the purpose of which is to catch bacteria and other irritants so they can then be coughed out of the lungs and an infection does it start in CF patients and mucous lining is much thicker which means it does what it's supposed to and catching the bacteria but it keeps them in the lungs the very thing which is intended to stop an infection becomes a breeding ground for those same bacteria over time the thickened mucous clogs the airways information spreads throughout the lungs and at 80% of CF patients this weeds to death by lung failure now there are two reasons why I picked CF as my example for how we can use bio dramatics the first on a more optimistic note that I'll get to in just a minute is that the past few years have seen some major advances and how cystic fibrosis is treated and those advances were based upon a deep understanding of how the mutated CF t our gene works ii however is it what I'm standing up here in front of all of you talking about bioinformatics and disease research do not think that this is just some sort of interesting intellectual exercise that you'll hear and go home forget about it's not for me and it shouldn't be for you either because I am speaking to you first and foremost not as a statistics major and first and foremost not as someone who wants to study bioinformatics in graduate school but if someone who has cystic fibrosis and on my behalf and on behalf of the other 100,000 people worldwide with CF I am Telling You our lives depend on our ability to better understand how the mutated gene works prior to about four years ago all the existing treatments for CF could be boiled down to two things queering mucus out of the lungs and killing the bacteria that had gotten into the lungs these were both helpful but neither of them actually addressed the root cause of the condition which is the thickened mucus thickened mucus and CF lungs is caused by defective chloride channels in the lung cells and so if you can imagine cyst fibrosis is like a giant hole in your bathtub wall with water gushing out of it everything we've had so far it's basically been ways of just devising larger buckets to scoop the water out faster so the tub doesn't flood this works for a little while it's been effective in bringing up the CF lifespan from about ten years in 1972 37 years today but as long as the water is still gushing freely out of the wall as long as the chloride shells are still malfunctioning the problem isn't really fixed in the last four years two new CF drugs Kuwait echo and or Camby have been developed that partially repair the defective core ID channels in the cells they're not cures but they're the first step towards patching over the hole and the only way we were ever able to develop such drugs was my understanding we intricately how the mutated gene works the decoding of hieroglyphics was no doubt an important milestone and our understanding of ancient Egypt but however fascinating such an intellectual pursuit may be it does not begin to compare in importance in urgency to our race to translate genetic information into the material for new disease treatments lives are being lost every day to heritable and genetically influenced diseases while we possess the information needed to save them walked away right in front of us if we're to combat the worldwide rise in cancer incidence the untold suffering in the global South from numerous virally cause tropical illnesses and conditions like cystic fibrosis which have destroyed millions of lives through its long history then we must continue to expand our genetic rosetta stone and we must use that stone to better understand diseases at the root causes I challenge my generation to be one that will be remembered as one which made full use of its mathematical and biological advances to save lives and in suffering the Egyptian culture which gave us the hieroglyphics has passed into history and all that remains of their civilization are the monuments they have left to us the day will come when we to wipe the Egyptians are nothing but a thing of the past to those who come after and they will know that we were the first generation and trusted with bioinformatics with all its potential for human health and when that day comes when we are nothing but a memory to those who come after when all that is left of us on earth is what we have done with this be our monument to them that we pushed medicine into an undreamt of frontiers that we left behind a better understanding of disease and had ever been possible before that we took up every weapon bioinformatics has given us and continue this fight against human disease thank you you

Channel: TEDx Talks

Views: 89,958

Rating: 4.9465585 out of 5

Keywords: TEDxTalks, English, United States, Science (hard), Biotech, Curiosity, Disease, Research

Id: _eHz6qzTCfc

Length: 11min 12sec (672 seconds)

Published: Fri Apr 01 2016