271

 

How Can I Evaluate Omics Data?

In the book, we learned about some omics techniques, such as genomics, transcriptomics,

or proteomics, and how they are related. It is important to know that systems analysis is

not so easy to formalize. However, it is easy to recognize important ingredients about

biological systems if you have enough biological knowledge. In practice, however, one is

mostly occupied with collecting and evaluating omics data (e.g. own experiments or from

databases, such as GEO). In most cases, the statistical software R is used, which allows

analysis and graphical representation and is also widely used, e.g. with the Bioconductor

tool for high-throughput data analysis (see Sect. 19.6). There are numerous online tutorials

and already prescribed scripts, it is best to simply go to https://www.r-­project.org/ and

https://www.bioconductor.org/ for information. In addition, several genome analysis pipe­

lines exist, e.g. GensearchNGS, in which we collaborate (Wolf B, Kuonen P, Dandekar T

et al (2015) DNAseq workflow in a diagnostic context and an example of a user friendly

implementation. Biomed Res Int 2015:403–497. https://doi.org/10.1155/2015/403497).

For proteome and transcriptome, our two papers Stojanović SD, Fuchs M, Fiedler J et al.

(2020) Comprehensive bioinformatics identifies key microRNA players in ATG7-deficient

lung fibroblasts. Int J Mol Sci 21(11):4126. https://doi.org/10.3390/ijms21114126) and

Fuchs M, Kreutzer FP, Kapsner LA et al (2020) Integrative bioinformatic analyses of

global transcriptome data decipher novel molecular insights into cardiac anti-fibrotic ther­

apies. Int J Mol Sci 21(13):4727. https://doi.org/10.3390/ijms21134727) provide a good

overview. For this it is best to look at the publication, there you will find instructions and

you can practice yourself.

If you want to look a little more into machine learning, you can check out our analysis

pipeline for diagnostic and prognostic signatures (Vey J, Kapsner LA, Fuchs M et  al

(2019) A toolbox for functional analysis and the systematic identification of diagnostic

and prognostic gene expression signatures combining meta-analysis and machine learn­

ing. Cancers [Basel], 11(10). pii: E1606. https://doi.org/10.3390/cancers11101606). A

nice application example is also shown in the paper Schweitzer S, Kunz M, Kurlbaum M

et al (2019) Plasma steroid metabolome profiling for the diagnosis of adrenocortical car­

cinoma. Eur J Endocrinol 180(2):117–125. https://doi.org/10.1530/EJE-­18-­0782).

19.1  Genomic Data: From Sequence to Structure and Function