244

of useful products such as citric acid or insulin. Finally, we have seen that regulation in

cells is very logical. If we want to recreate this with a computer, we first have to find out

which protein interacts with which. This can be done using protein interaction databases

such as STRING and KEGG. Then you can assemble the logical (Boolean) network into

a regulatory circuit in the cell. If this is made computer-readable (e.g., XML format), a

program can then simulate the dynamics of regulation without having to know exactly

how fast the process takes place. For this, the model is then only “semi-quantitative”, i.e.,

it only tells what comes sooner or later, what is more or less activated in the cell.

Nevertheless, this helps to describe how our nervous system works or to find new drugs

against heart failure.

Principles for understanding bioinformatics and modern biology we have learned in

the second part. Heuristic, i.e., fast, but not completely accurate searches speed up mod­

ern bioinformatics programs. Bioinformatics decodes coded information in cells, and liv­

ing cells use different codes and levels of language. The analysis tasks for the computer

are either easy, meaning that in the foreseeable future the computer can handle them, or

they are unpredictably long. This happens especially easily when many combinations

are tested.

Every organism is a complex system, but they behave in fundamentally similar ways.

One can infer their behavior through big data, such as omics techniques. Emergence is at

the heart of this, i.e., the appearance of completely new system properties as components

come together, especially in human consciousness when a critical number of neurons

come together in a previously unfounded way. Modular structure, positive and negative

feedback loops are basic properties. A basic theme of biology is the consideration of evolu­

tion. Bioinformatics likes to use phylogenetic trees and evolutionary comparisons to

quickly identify basic properties (many organisms; conserved regions in a protein

sequence) and specific properties.

All these data analyses allow detailed insights into the molecular biology of the cell by

looking step by step at involved protein sequences, their localization in the cell and their

properties.

Bioinformatics becomes fascinating (Part 3) when data analysis provides surprising

biological insights. For example, modern genome analysis makes it figuratively clear that

genetic modifications affect every human genome and that everyone carries “good” and

“bad” genes with them, useful or harmful depending on the environment. Molecular

sequences only ever make sense in the context of the cell. Understanding this language can

be used for synthetic biology and protein design, such as the nanocellulose chip.

Bioinformatics is also making the powerful structure of our brains and limitations for

natural and artificial intelligence clearer. Bioinformatic modelling and simulations also

encompass ecosystems, infections (currently: Covid19 pandemic) and climate models, the

internet and sharpen the view for chances and problems of our digital society.

Conclusion  Bioinformatics uses computers to better understand biological problems,

i.e., to find similarities between molecules, for example at the sequence level. It can break

17  Conclusion and Summary