In my opinion, we are witnessing the growing of using bioinformatics for curing various types of diseases and in the near future it will be extended in all different aspects even nonmedical applications. What is your idea?
There was a shift in lifescience thinking around the year 2000. This was summarized in a quote from a European biotech conference that I read in a Journal once, "In 10 years time those doing hypothesis-driven experimentation will be like those that still believe in flat earth theory." WRONG! The basic problem in bioinformatics isn't the math and statistics, it's that "no one can afford to run all the experiments required to reach statistically-valid conclusions" (Steve Naylor, former CSO of Beyond Genomics, now defunct). Basic chemimetrics says you need a minimum of 3 independent samples per variable being measured, more if there is error in the measurement. When you are looking at up and down regulation of a 100,000 genes, proteins, and metabolites, that means a minimum of 300,000 samples. Most people can afford just 12-24, so all this hypothesis-free nonsense has resulted in nothing in 2 decades. On the plus side, you don't need any active brain cells to do hypothesis-free experimentation, so it's perfect for today's PhD students. At least that is my considered opinion.
My short answer is that the future of bioinformatics is "Big deal". As the cost of producing sequence data and genotyped data is becoming dramatically cheaper due to the rapid and relentless technological advancement, an equal match in new and more efficient bioinformatic tools are required to keep with the analysis. Indeed, this is already happening in human precision medicine, where amazing sucess have already been obtain in novel gene editing therapy.
@ Mohammed,
I think we have great improvements on computer science algorithms to deal with Big Data, e.g. deep learning and there will not any problem for bioinformatics later. Thus, what are different applications would be used bioinformatics in the future due to the improvements of computer science simultaneously?
The Big Data problem has already been tackled in other industries. The real issue is what are the questions (hypotheses) and can we differentiate competing hypotheses with the data we have or can affordably generate. That requires mathematical models of the biology, something which biologists are poorly equipped to do.
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Bioinformatics