S2: Standardization of Bioinformatics

Standardization of bioinformatics analysis workflows for omics-based safety assessment
Dr Pekka Kohonen, Karolinska Institutet

Pekka Kohonen


Institute of Environmental Medicine,
Karolinska Institutet, Stockholm, Sweden


Pekka Kohonen, Penny Nymark and Roland Grafström


Omics analyses provide a potentially comprehensive means assessing signaling pathway and gene expression processes that are active and changed as a result of diverse treatments.

Encompassing typical data streams like RNA-seq and microarrays, data from these methods needs to be pre-processed, normalized, scored for biological activities, interpreted and placed in a biological and risk assessment context.

There is extensive bioinformatics literature with conventions for analyzing each data type which need to be adapted for the purposes of regulatory toxicology.

Reproducible analysis is facilitated by programmatic or scripted data analysis in general and by the use of the many tools available via the R statistical programming language and the R/Bioconductor bioinformatics suite (over 1000 packages). From a bioinformatics perspective, most types of analysis and packages can be assessed for their potential adaption and application.

Standardization of pathway analysis methods is perhaps the most urgent need to regulatory application, and benchmark dose analysis might require further developments. Establishing practices for open code and data sharing is a further point.

The complexity of bioinformatics standardization issues will be exemplified, and tentative solutions discussed.