SOHPIE

SOHPIE performs multivariable differential co-abundance network (DN) analyses to identify taxa that are differentially connected between sample groups while adjusting for covariates in microbiome studies.

Key Features:

• Covariate Adjustment: Implements a regression-based approach to adjust for additional covariates (e.g., age, diet, medication) when identifying differential connections among taxa.
• Multivariable Analysis: Supports multivariable DN analyses that consider multiple influencing factors simultaneously.
• Pseudo-value Estimation: Leverages pseudo-value information to estimate differential networks.
• Comparative Capability: Enables identification of differentially connected (DC) taxa between clinical or phenotypic groups and provides regression-based covariate adjustment not present in MDiNE or NetCoMi.

Scientific Applications:

• Differential Connectivity Analysis: Identify taxa that are differentially connected between individuals with varying clinical and phenotypic characteristics.
• Host–Microbiome Association Studies: Investigate microbial community interactions and their associations with host characteristics while adjusting for confounders such as age, diet, and medication.

Methodology:

SOHPIE applies a regression-based statistical approach leveraging pseudo-value information to estimate multivariable differential co-abundance networks and adjust for covariates.