11
$\begingroup$

What was the first piece of work in computational biology? I'm ideally looking for a paper.

I am not interested in works that involve data management or data analysis but work that model biological processes through numerical simulations or numerical approximations of analytical results.

$\endgroup$
  • 1
    $\begingroup$ As well described in nature.com/nrg/journal/v1/n3/full/nrg1200_231a.html computational biology really grew out of protein structure determination. $\endgroup$ – DrDanielSwan Aug 4 '16 at 14:56
  • 3
    $\begingroup$ for my money it's the Lotka–Volterra predator–prey model first expounded in 1910, and extensively analyzed by 1925. en.wikipedia.org/wiki/Lotka%E2%80%93Volterra_equations $\endgroup$ – RBarryYoung Aug 4 '16 at 20:02
  • 1
    $\begingroup$ You might also be interested in this discussion on www.biostars.org: The Oldest Bioinformatics Publication $\endgroup$ – user1136 Aug 4 '16 at 23:31
  • 1
    $\begingroup$ @RBarryYoung The simple Lotka-Volterra equations can be analyzed analytically (see this post), no need for numerical simulations. $\endgroup$ – Remi.b Aug 5 '16 at 3:29
  • 1
    $\begingroup$ Does phylogeny inference qualify ? $\endgroup$ – bli Aug 5 '16 at 8:26
8
$\begingroup$

Another nomination, if you include infectious disease epidemiology as part of biology and hence computational simulations of epidemics as part of computational biology:

Measles periodicity and Community Size, M. S. Bartlett, J. Roy. Stat Soc. A, 120(1), 1957.

The computations were run on the Manchester computer. Possibly the most entertaining part of the paper is the discussion afterwards from one of the computing assistants:

Mr. J. C. GOWER: I should like to describe in a little more detail the programme for the Manchester computer which has produced the results that Professor Bartlett has been discussing ... Owing to the fact that the computer makes not infrequent mistakes and in view of the apparent impossibility of getting an overall check ... it is necessary to repeat the calculations ... The random numbers are produced in batches of 64. Each batch is tested for divergence from the expected number of unit digits. If the test fails a new batch is produced and tested. If three successive batches fail the machine stops and hoots continuously.

Only once in the sixteen months during which the programme has been running have three successive batches failed ...

But the winner (also in population biology) might be the one linked in the comments to a Biostars discussion, Gene frequencies in a cline determined by selection and diffusion R.A. Fisher Biometrics 1950. On p. 169 the author says

I owe this tabulation to Dr. M. V. Wilkes and Mr. D. J. Wheeler, operating the EDSAC electronic computer

(the tabulation is the solution of the differential equation $\frac{d^2 q}{dx^2} = 4 x (1-q)q$ with boundary conditions $q=1/2$ at $x=0$ and $q=0$ as $x \to \infty$); the Wikipedia page about EDSAC (linked above) claims

[Fisher's study] represents the first use of a computer for a problem in the field of biology

$\endgroup$
  • $\begingroup$ Welcome to BioSE and really nice to see such a prominent researcher as a contributor. $\endgroup$ – fileunderwater Aug 22 '16 at 23:08
8
$\begingroup$

I don't believe you'll ever find the first work in bioinformatics (or computational biology, as you put it), however the field really began in the times of accumulating data about protein biochemistry. Computational biologists (before they had access to the computer) would be writing and analyzing morphologies and types of proteins with pencil and paper.

But you can go even further back than this. "Alignment" is another old technique used in bioinformatics. This is establishing the amount of similarity between two DNA sequences, or rather the degree of similarity between any two objects or data sets in computational biology.

Your question, though, asks for a scientific paper (one of the oldest) on bioinformatics. This is arguably M. O. Dayhoff's paper: "A Computer Program to Aid Primary Protein Structure Determination." (1962)

This is the first formal paper that limits the scope of computational biology to the definition provided by this BioPlanet article and generally agreed by many:

Bioinformatics is the application of computer technology to the management of biological information.

$\endgroup$
3
$\begingroup$

Depending on your definition of computational biology (e.g. bioinformatics and mathematical modelling can be quite different), there is an oft-cited reference to Alan Turing's work after the second world war, modelling what he called "morphogens" in the emergence of mathematical patterns in nature. As I understand it he modelled diffusion of antagonistic and complementary hypothetical molecules in space and time. I'm fairly sure this is one of the first examples of using the very early computers, while they were still reminiscent of counting machines for biological problems.

I don't have an exact reference for you right now, but some google-fu of "Turing" and "morphogens" will drag something up.

EDIT: here's the paper, 1952 is the year to beat!:

http://www.dna.caltech.edu/courses/cs191/paperscs191/turing.pdf

$\endgroup$
  • 1
    $\begingroup$ Please note that - while Turing is famous, and he usually gets most credit - his (absolutely great) work was lacking multiple important pieces - and was not necessarily computational biology - and had very little influence on biological research. "Turing's morphogens" were discovered independently, with the inclusion of missing pieces, and shown computationally, by Alfred Gierer and Hans Meinhardt ( for historical perspective: dev.biologists.org/content/develop/early/2016/03/22/… ) $\endgroup$ – tsttst Aug 5 '16 at 15:44
  • 1
    $\begingroup$ yes, but in the paper Turing does say (p. 65) "The figures in table 1 were mainly obtained with the aid of the Manchester University Computer." But the precedence probably has to go to the Fisher 1950 population genetics paper discussed in the Biostars thread mentioned in the comments above, Gene frequencies in a cline determined by selection and diffusion/1/238.pdf $\endgroup$ – Ben Bolker Aug 6 '16 at 1:05
2
$\begingroup$

I don't have any idea about the first paper in computation biology (interpreted by me as papers that use computers and computer simulations to solve biological problems). However, some population ecologists were early in adopting computer simulations to solve population models. Your should look at Micheal Hassell, which I know was early in using computer simulations. One of his first papers is New inductive population model for insect parasites and its bearing on biological control. from 1969 (pdf), where models are simulated on the "Oxford KDF9 computer". His other papers and his collaborators could be a starting point.

$\endgroup$
1
$\begingroup$

There is some ambiguity in words; however "computational biology" is often used to stress that work extends beyond sequence information (and sequence derived properties) and data management (which would rather be described by "bioinformatics").

If the question is about the first work that used computer simulations, and algorithms to discover and mechanistically explain some complex biological problem, the work of Hans Meinhardt and colleagues comes very close to a first piece of computational biology.

For instance they discovered how patterns could form; Gierer et Meinhadt 1972, and other references contained in his obituary

$\endgroup$
  • $\begingroup$ Looks like an interesting author to consider. Could you give more info of the years of various publication and the type of work he did ("discovering the formation of patterns" is very vague). $\endgroup$ – Remi.b Aug 4 '16 at 15:17
  • $\begingroup$ Have you tried pubmed ? $\endgroup$ – diffracteD Aug 4 '16 at 16:23

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.