Identifying correlations driven by influential observations in large datasets

Abstract

Although high-throughput data allow researchers to interrogate thousands of variables simultaneously, it can also introduce a significant number of spurious results. Here we demonstrate that correlation analysis of large datasets can yield numerous false positives due to the presence of outliers that canonical methods fail to identify. We present Correlations Under The InfluencE (CUTIE), an open-source jackknifing-based method to detect such cases with both parametric and non-parametric correlation measures, and which can also uniquely rescue correlations not originally deemed significant or with incorrect sign. Our approach can additionally be used to identify variables or samples that induce these false correlations in high proportion. A meta-analysis of various omics datasets using CUTIE reveals that this issue is pervasive across different domains, although microbiome data are particularly susceptible to it. Although the significance of a correlation eventually depends on the thresholds used, our approach provides an efficient way to automatically identify those that warrant closer examination in very large datasets.

Oral vancomycin treatment suppresses gut trypsin activity and preserves intestinal barrier function during EAE

“…To identify gut commensals that are associated with gut trypsin activity and EAE severity, we performed correlations under the influence (CUTIE) 44 correlation analysis on all the mice. We found that four Lactobacillus species including L. reuteri , L. frumenti , L. vaginalis , and L. f/r/a/o/p/v negatively correlated with trypsin activity and EAE severity during the pre-symptomatic phase of the disease ( Figures 4 A and 4C).…”

“… 71 http://rdp.cme.msu.edu ; RRID: SCR_022773 Resphera Insight (v2.2) Resphera Biosciences Shaikh et al., 2021 65 https://respherabio.com/ CUTIE (Correlations Under The InfluencE) Bu et al. 44 https://github.com/clemente-lab/CUTIE.git ImageJ Schneider et al. 72 https://imagej.net/ij/index.html ; RRID: SCR_003070 Prism v9.3.1 Graphpad Software https://www.graphpad.com/ ; RRID: SCR_002798 Other Illumina Inc. https://www.illumina.com/ …”

“…CUTIE correlation analysis, which significantly reduces false positives while increasing sensitivity against false negatives, was used to identify significant correlations between relative abundance of bacterial taxa and fecal trypsin activity and cumulative EAE clinical score. 44 …”

Bad company? The pericardium microbiome in people investigated for tuberculosis pericarditis in an HIV-prevalent setting

“…Correlation analysis with CRP and microbiome: Only sequence data with paired CRP measurements were included. Microbial relative abundances were correlated with CRP measurements using Correlations Under the Influence (CUTIE) 28 .…”