I plan to apply for PhD candidate or master student in bioinformatics field.
I wonder what the big questions in bioinformatics field nowadays are.
I want to do something which maybe help push the bioinformatics forward. Please give me some clues or tips.
Broadly speaking, biggest questions in bioinformatics are the same as biggest questions in biology. That's because bioinformatics ultimately solves biological problems. The difference is in techniques used, so technical aspects of difficulty will be different. When you study cancer in mice as a biologist, there are some problems regarding mice maintenance, microscopy training, reproducibility of diagnosis, etc. As a bioinformatician studying the same field, the problems would be in the areas of image storage, noise removal, machine learning, etc.
Just so the answer is not completely general, here is what various smart people think about this problem:
https://www.ncbi.nlm.nih.gov/books/NBK25461/
I am partial to what is described in B.6, but all are valid points of view. It is fairly easy to guess from the answers whether the background of respondents is on the wet-lab or computer side.
This'll be a quick response since this has been answered on the forum before.
Broadly speaking, there are lots of questions left in terms of generating faster code and more efficient datastructures. As the size and scale of bioinformatics data is rapidly beginning to outpace even the likes of CERN, having code 'tricks' (recent examples that spring to mind might be sketches/hashing), and more efficient databases etc., will be paramount. Even more so when it comes to training AI/ML tools (which like it or not, aren't going away any time soon).
The field is too big to give you specific questions. You need to look in to areas than interest you more closely, or see what is on offer for PhD projects at various institutions.
In short, if you can't see the questions yet, you need to look more closely.
Instead of answering your question I will try to give a list of things to avoid:
To push the bioinformatics forward, please learn and improve the areas e.g., programming/algorithmic skills, understanding molecular biology and the ability to understand research papers.
Bioinformatics is a big field and there are many problems to work on. If you can keep up with the current literatures and methods, you will see "what the big questions".
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
"IS IT ZERO OR ONE-BASED ?" :-)
-1A problem for clinical bioinformaticians: 3) Can I get first authorship considering I am not an MD?
Personally I think that bioinformatics will be largely pushed forward after solving this problem:
https://en.wikipedia.org/wiki/P_versus_NP_problem
(discussed here https://biology.stackexchange.com/questions/16621/list-of-np-hard-problems-in-biology-bioinformatics )
You mention P versus NP problem. Is it an even too big question for bioinformatics which is a specialized field other than computer science or math science? Are there any specific topics in bioinformatics and more related to biology or human? Maybe I am too "low" for you. Anyway, thank you for your response.
You should add some details. Bioinformatics is an extensive field. Are you more into data analysis, so want to work ona biological problem or are you more a programmer and want to develop a new tool, or rather very much into developing new analysis strategies?
I am very glad that you can question me. I want to share some feeling with you guys. I cannot see the questions or difficulties in bioinformatics. Can you feel what I am talking? For example, if someone want to get a whole genome sequence of some organism, he can actually get it, though it cost some more money or time. I want to know what the real questions the scientists are tackling with in bioinformatics. Maybe I am too little experienced, so I look like a fool.
Well, that shows how much you know about the field.
I understand the field is quite vast for a beginner.
I suggest you check on some "big shots" in the field and their research topics, usually you can find their current interests on their websites. Being not academic, my interests are far less basic, so I'm not a big help in identification of those key persons, but there's some institutes like EBI, EMBL, NCBI, JGI, etc. or journals like Nature Methods (filter on bioinformatics), Bioinformatics.
In the end, the importance of a (research) question can be relatively subjective and/or a matter of fluctuating trends, with the exceptions of some really fundamental problems, like for example German's example above...
if it would only be that simple!
But unfortunately it is not, which also kinda illustrates that a decent background in the topic/field is crucial to ask the "correct/good" questions.
No, I also provided the list of topics that will largely benefit from solving this problem (such as multiple sequence alignment or genome assembly). It is directly related to biology and for humans too. Also Network Inference (finding relationships between genes) is a NP-hard problem. Non-Negative Matrix Factorization is, I believe, NP-hard, and it is widely used in e.g. cancer signatures determination. Drug design and protein modelling is NP-hard.
This problem is indeed big, but that's what you've asked.
If the problem is too big for you, you may concentrate on finding approximate solutions for NP-hard problems in reasonable time.