Mistakes in the re-analysis of lipidomic data obtained from a human model of resolving inflammation lead to erroneous conclusions

Abstract

AbstractRecent years have seen an increased interest in the biology of specialized pro-resolving lipid mediators (SPM) with many investigators evaluating both their endogenous production as well as their biological and pharmacological properties. This increased interest has led to a rapid evolution in our understanding of both the biological and pharmacological activities of these mediators with their endogenous formation and biological activities being documented in a wide range of species that spans the evolutionary tree including fish, planaria and humans. Despite this plethora of evidence in a recent article Homer and colleagues claim that the reanalysis of a published dataset - partly originating from our laboratory - demonstrates that these molecules are absent in human blister exudates, contrary to the evidence published earlier. Based on their observations these Authors conclude that SPM are not relevant to human resolution biology. In this manuscript a reanalysed the dataset using unbiased methodologies and criteria that align with those recently proposed by the UK Consortium on Metabolic Phenotyping was performed together with the re-analysis of the dataset using criteria described by Homer and colleagues. Results from this re-analysis confirmed presence of SPM in human blister exudates and that the methodologies employed for quantitation of these molecules are robust. We also discuss how the results obtained in the article published by Homer and colleagues present several points of concern including the use of employ an arbitrary cut-off value to assign the noise for all the transitions used that does not take into consideration the fluctuation of the signal in each transition and therefore is not truly representative of the background signal. The use of different transitions to those employed in the original analyses and misreporting of findings based on the criteria employed. In conclusion the evidence presented herein demonstrates that correct application of rigorous criteria accepted by the community is essential in ensuring accurate identification of mediators and avoid blatant mistakes which can impact on the scientific development of the field.