This manuscript tests the accuracy of correlation-based inference methods for inferring interactions between viruses and microbes. Interaction networks are generated a priori using BiMat [1], and community dynamics are simulated in silico using a system of differential equations. The correlation-based inference methods tested include: standard correlation (Pearson, Spearman, and Kendall), time-delayed correlation (Pearson, Spearman, and Kendall), Extended Local Similarity Analysis (eLSA) [2], and Sparse Correlations for Compositional Data (SparCC) [3].
Downloads and installation instructions for [1], [2], and [3] can be found here:
BiMat (Matlab): https://bimat.github.io
eLSA (Python): https://bitbucket.org/charade/elsa/wiki/Home
SparCC (Python): https://bitbucket.org/yonatanf/sparcc
A preprint of the manuscript can be found on BioRxiv: https://doi.org/10.1101/176628.
This code is archived on Zenodo:
[1]: Flores et al (2016), BiMat: a MATLAB package to facilitate the analysis of bipartite networks. Methods Ecol Evol, 7: 127-132. https://doi.org/10.1111/2041-210X.12458.
[2]: Xia et al (2011), Extended local similarity analysis (eLSA) of microbial community and other time series data with replicates. BMC Systems Biology, 5(Suppl 2), S15. https://doi.org/10.1186/1752-0509-5-S2-S15.
[3]: Friedman and Alm (2012) Inferring Correlation Networks from Genomic Survey Data. PLOS Computational Biology 8(9): e1002687. https://doi.org/10.1371/journal.pcbi.1002687