How Far Does Bioinformatics Go?

Question

[Egon Willighagen]

Being a metabolomics, and drug discovery dude, I consider myself a bioinformatician (well, I also consider myself a chemist, cheminformatician, statistician, and chemometrician, but that's not relevant to my question).

However, some peers see bioinformatics restricted to stuff to do with DNA sequences, that is genomics. So, from a historical and literature perspective, what is bioinformatics? Please do back up your answer and argument with citations to primary literature.

[Jeroen Van Goey]

Since you asked for citations to primary literature, I recently came across this article: Earliest pages of bioinformatics.

This review is a brief outline of the chronology and essence of early events in bioinformatics, covering the period from 1869 (discovery of DNA by Miescher) to 1980-1981 (beginning of massive sequencing). For the purpose of this review, bioinformatics is understood as a chapter of molecular biology dealing with the amino acid and nucleotide sequences and with the information they carry.

Answer 1

[Jeroen Van Goey]

Bioinformatics is the field of science in which biology and computer science/information technology merge into a single discipline. The ultimate goal of the field is to enable the discovery of new biological insights as well as to create a global perspective from which unifying principles in biology can be discerned.

Listed below are some of the major events in bioinformatics over the last several decades. Most of the events in the list occurred long before the term, "bioinformatics", was coined.

I've tagged each entry with either:

• (BIO) if it was an event which was predominantly important in the field of biology.
• (IT) if it was an event which was predominantly important in the field of computer science/information technology
• (BIOINFO) if it was an event where biology and computer science/information technology truly merged and we can really speak from bioinformatics.

As you will notice, it becomes increasingly difficult/subjective to catalogue events with exclusively one tag, so the main point to take away is that in the course of bioinformatics history there has been a constant exchange of ideas between biology, computer science/information technology and bioinformatics.

Compiled from different sources, including:

• A Short History of Bioinformatics, by Allen B. Richon
• History of Bioinformatics
• Bioinformatics Milestones
• History, Origin & Bioinformatics Events

[Daniel]

This is awesome. now where is the 2004 to present... :-)

[Istvan Albert]

Wow. That's a pretty good history lesson! Thanks!

[Egon Willighagen]

So, according to this list Bioinfo is genome only, right?

Answer 2

[Nicojo]

As much as I agree with the impressive list from BioGeek, I have to say that it is a non-exhaustive, genomics centric list.

If we look at the first statement he mentions: "Bioinformatics is the field of science in which biology and computer science/information technology merge into a single discipline."

From that statement I understand that anything "biological" studied using a "computer" should have its place in what we call "Bioinformatics".

One perfect example would be all those people working on proteins and not DNA/RNA. I wouldn't say that those are the same field in Bioinformatics, however you may argue that they fit together... As an example of a "Bioinformatics Center" that focuses on proteins I would give: Stockholm Bioinformatics Center

Another example would be those people trying to understand how molecules diffuse within the cell cytoplasm. That is a lot of computer work that's directly looking at understanding a phenomenon in a biological context. This type of project is bordering on many disciplines and not just biology and informatics, but also physics. Nevertheless, shouldn't that be a "bioinformatics" discipline too? In this category I would give the Biomatter @ MOSAIC ETH Zurich

What about Systems Biology (even if they don't want to be called bioinformaticians), shouldn't they be called bioinformaticians too?

And finally (although far from exhaustive) I'll give one last example of something I consider bioinformatics: the E-Cell Project

I hope this answers your question! In my opinion, bioinformatics is NOT only "genome stuff" and I would extend it to yourself too ;)

Answer 3

[Giovanni M Dall'Olio]

Many people consider that bioinformatics began with the work of Margaret Dayhoff and the Pam matrixes. She compiled the first collection of protein sequences available at the time, publishing the Atlas of Protein Sequences and Structure, and she developed the first method to give a score to the similarity of two proteins, the PAM matrix.

For me, bioinformatics is everything that derived from Margaret Dayhoff's work. Compiling data and organizing it, developing tools to compare and handle information, share the data with other people: if you read her biography you will find everything already there.

About the modern bioinformatics, I like to think of it as the science of doing experiments or part of them using computers at least for some steps. I like to think that there is no difference between the work in a wet lab and that in front of computer: when you are planning a bioinformatics project, you also have to think of an hypothesis, on how to verify it and on which tests and controls you will use. This is probably something that many people didn't understand yet, as they think that bioinformatics is just 'writing programs' and the don't even know what a test is and how much time it takes to write a program.

Answer 4

[Patrick Koks]

Of course, it's not only about definitions and terminology, but in this essay, Pauline Hogeweg, who coined the term 'bioinformatics' in 1978 or 1980 or as early even as 1970, describes what her group initially used the term for: "study of informatic processes in biotic systems". Informatics or informatic processes are actually pointing at the 'flow of information through a living cell, individual, species or evolution'. Hogewegs' work is focused on pattern recognition, morphogenesis and evolution.

In other words, bioinformatics covers a very broad range of topics and originally wasn't even meant for sequence based , or even hard data-driven research.

So metabolomics easily fits in this definition, as long as you study samples from a 'biotic' origin. Reasoning that (small) metabolites have an important role in information transfer is obvious, certainly in drugs discovery.

Does this make you a happier bioinformatician?

I hope not. (as your own opinion and feeling is what matters in the first place)

Is this answer placing metabolomics right at the heart of the bioinformatics community?

I think not, but the good news is that after having sequenced 'everything' in 2012 or 2013, and some chewing on this in the years after that, bioinformaticians will have to follow 'experimentalists' to the core discipline of biology metabolomics is to become sooner or later.