93
8
When Does the Computer Stop Calculating?
Abstract
The question of when a bioinformatics problem will be completed is difficult to answer
for problems with built-in combinatorics. Alan Turing generally modeled all comput
able problems using the Turing machine, an idealized abstract computer. All non-Turing
computable problems cannot be solved by computers and remain tasks for humans.
Many particularly interesting problems in bioinformatics are NP (nondeterministic
polynomial complexity) problems, such as protein structure prediction and most net
work and signal computation or image processing. In general, more powerful comput
ers, the bundling of many computer nodes (parallelisation) and application-specific
chips can also directly increase computer performance, for example with omics data.
We remember that bioinformatics analyses biological data with programs (Sect. 2.1), col
lects them in databases (Sect. 2.2) and then maps the biological relationships in models.
But how good are bioinformatic models? Well, bioinformatics tries to use computers to
make “good” and comprehensible biology. One can have fundamental reservations about
this. After all, life is a quality rather than a quantity. Experiences are not seldom simply
indescribable, and also a bacterium or also your own mind and even the brain are not sim
ply a kind of chip (bacterium) or supercomputer (we ourselves). We are infinitely much
more, and who cannot understand this at all, should now go to a good theater play (no
cinema effect, it is better to experience this “live”) or talk for a few minutes with a patient
in a psychiatric ward, then may be he will better fathom what we want to say.
© Springer-Verlag GmbH Germany, part of Springer Nature 2023
T. Dandekar, M. Kunz, Bioinformatics,
https://doi.org/10.1007/978-3-662-65036-3_8