micInt-package: micInt: Find microbial interactions

In AlmaasLab/micInt: Find microbial interactions

micInt: Find microbial interactions

Description

This packages utilizes different approaches to find and analyze microbial interactions. The CoNet approach utilzes similarity measures and bootstrapping-permutation methology to find significant correlations between different OTUs in a microbiome data set. The analysis are automized and run in parallel for convenient comparison of different similarity measures and methods. The Lotka-Volterra approach is based on time series and estimates the Lotka-Volterra coefficients.

Details

Common

The starting point for this packages is an OTU table. See description of its format under runAnalysis. Note the functions OTU_stats and refine_data which are used internally by many of the functions, but can be useful in user-written code as well.

CoNet

For the CoNet approach, the major function is runAnalysis which provides a more or less complete pipeline. This is a wrapper around the package ccrepe. For more advanced users, the lower level API using ccrepe_analysis might be preferable. In addition, there are many auxillary functions such as generating interaction tables (create_interaction_table), writing them to file (write.interaction_table) and making diagnostic plots of the tables (autoplot.interaction_table).

Lotka-Volterra

This approach is based on The Lotka-Volterra systems are generated by integralSystem For the time being, ridge_fit is the only function available to solve the systems (works even if they are over- or underdetermined) by ridge regularization. The regularization parameters have to be determined somehow, this is where cv.LV comes into play. It performs cross-validation in order to pick the best regularization parameters. Finally, the estimated coefficients can be put into the differential equations again in order to predict future dynamics of a microbial community using predict.LV

Author(s)

Maintainer: Jakob Peder Pettersen [e-mail] (ORCID)

Other contributors:

• Emma Schwager [contributor]

• Craig Bielski [contributor]

• George Weingart [contributor]

References

Faust K and Raes J. CoNet app: inference of biological association networks using Cytoscape F1000Research 2016, 5:1519 (doi: https://doi.org/10.12688/f1000research.9050.2)

Karoline Faust et al. "Microbial Co-occurrence Relationships in the Human Microbiome". In: PLoS Comput. Biol. 8.7 (July 2012). issn: 1553-734X. doi: https://doi.org/10.1371/journal.pcbi.1002606.

Richard R. Stein et al. “Ecological Modeling from Time-Series Inference: Insight into Dy- namics and Stability of Intestinal Microbiota.” In: PLoS Comput. Biol. 9.12 (desember 2013). issn: 1553-7358. doi: https://doi.org/10.1371/journal.pcbi.1003388.

P. H. Kloppers and J. C. Greeff. “Lotka-Volterra model parameter estimation using experiential data”. In: Appl. Math. Comput. 224 (Nov. 2013), pp. 817–825. ISSN: 0096-3003. DOI: https://doi.org/10.1016/j.amc.2013.08.093

See Also

refine_data Convert an OTU table into a refined table (required for some functions in the package)

OTU_stats Get statistics on OTUs

runAnalysis High level CoNet wrapper

ccrepe_analysis Low level CoNet wrapper

create_interaction_table Make CoNet interaction table

integralSystem Make equations for estimating the Lotka-Volterra coeffcients

ridge_fit Solve Lotka-Volterra systems by ridge regularization

cv.LV Find optimal regularization parameters by cross-validation

predict.LV Predict the development of a microbial community by estimated Lotka-Volterra coeffcients

AlmaasLab/micInt documentation built on April 1, 2022, 10:37 a.m.