294

https://pinguin.biologie.uni-­jena.de/bioinformatik/networks/) and YANAsquare (https://

www.bioinfo.biozentrum.uni-­wuerzburg.de/computing/yanasquare/) are recommended at

this point. Of course, one can look at enzymes in more detail, for example with the help of

metabolic control theory. A good introduction is the book by David Fell and Keith Snell

(1997), which shows how to calculate the strength with which an enzyme controls a meta­

bolic flux, regulatory coefficients and the like: Understanding the Control of Metabolism.

Also helpful is the book by Reinhart Heinrich and Stefan Schuster (1996) The Regulation

of Cellular Systems. More recent results can be found in numerous individual publications

(just browse the Internet yourself).

How Can I Better Understand Signal Cascades by Measuring the Encoded

Information?

Here we have learned about Shannon entropy. An encoding is done with bits, and there are

different levels of encoding. The paper by Heinrich et al. (2002) nicely translates the

signal-­to-noise problem into a biological application example, kinase signal cascade

(Heinrich R, Neel BG, Rapoport TA (2002) Mathematical models of protein kinase signal

transduction. Mol Cell 9(5):957–970).

The decoding of protein or nucleotide sequences using the genetic code is fast and reli­

able, but the other codes are much more difficult to decipher. For example, the three-­

dimensional structure is difficult to predict from the protein sequence (something for

specialists; the accuracy for the best methods [e.g. Zhang lab, QUARK server: https://

zhanglab.ccmb.med.umich.edu/QUARK/ as well as David Baker lab, Robetta:

https://robetta.bakerlab.org], if the structure is not too complex and unknown, is about 4–6

angstroms). Therefore, we focus more on 3-D predictions by homology modeling. True

3-D predictions for RNA (more degrees of freedom) are even more difficult. Sugar code

decoding is just beginning (https://www.ncbi.nlm.nih.gov/books/NBK1965/; NIH

Bookshelf Glycomics; Chauhan JS, Bhat AH, Raghava GP, Rao A (2012) GlycoPP: a

webserver for prediction of N- and O-glycosites in prokaryotic protein sequences. PLoS

One 7(7):e40155. https://doi.org/10.1371/journal.pone.0040155). And the lipid code is

even less understood.

Are There Also Problems for the Computer, and When Does It Become Difficult for

the Computer?

This is an exciting topic for computer scientists. In practice one should be careful to think

that there are simple general solutions how fast a computer will solve a given task. The

Wikipedia page (https://en.wikipedia.org/wiki/P_versus_NP_problem) about this is

already very instructive. But Gerhard J.  Woeginger’s page (https://www.win.tue.

nl/~gwoegi/P-­versus-­NP.htm) only opens the eyes how difficult, exciting and versatile this

seemingly simple topic is, especially in the formulation: If the solution to a problem is

easy to check for correctness, is the problem itself easy to solve? If so, all NP-problems are

convertible into P-problems; but probably this is not the case, or at least it has been stub­

bornly open as a question for decades.

19  Tutorial: An Overview of Important Databases and Programs