[Conference] Investigation of technical variance in microbiome study from varying analytic pipelines

Our new paper has been accepted in the international conference on Data and Text Mining in Biomedical Informatics (DTMBIO) 2023.

We investigated such technical variance due to heterogeneous analysis pipeline configurations by analyzing 16S ribosomal RNA sequencing data from the samples of human lower respiratory tract and environmental controls. Comparative assessment of various analysis pipeline configurations shows almost negligible difference. We also identified that using protected brush with bronchoscopy was more reliable than using conventional washing protocol, as their results showed less analytical variances. Our study suggests that using different analysis configurations can cause relatively small effects in results of 16S rRNA sequencing data analysis. Nevertheless, using different configurations for specific analysis components generally caused larger discrepancies than the other components. Thus it is necessary to configure analysis pipelines in consideration with such analytical variances.