LongDat: an R package for covariate-sensitive longitudinal analysis of high-dimensional data

Abstract

Summary We introduce LongDat, an R package that analyzes longitudinal multivariable (cohort) data while simultaneously accounting for a potentially large number of covariates. The primary use case is to differentiate direct from indirect effects of an intervention (or treatment) and to identify covariates (potential mechanistic intermediates) in longitudinal data. LongDat focuses on analyzing longitudinal microbiome data, but its usage can be expanded to other data types, such as binary, categorical, and continuous data. We tested and compared LongDat with other tools (i.e., MaAsLin2, ANCOM, lgpr, and ZIBR) on both simulated and real data. We showed that LongDat outperformed these tools in accuracy, runtime, and memory cost, especially when there were multiple covariates. The results indicate that the LongDat R package is a computationally efficient and low-memory-cost tool for longitudinal data with multiple covariates and facilitates robust biomarker searches in high-dimensional datasets. Availability and Implementation The R package LongDat is available on CRAN (https://cran.r-project.org/web/packages/LongDat/) and GitHub (https://github.com/CCY-dev/LongDat). Supplementary information Supplementary data are available at Bioinformatics Advances online.

Gut microbiome alterations precede graft rejection in kidney transplantation patients

“…We performed longitudinal analysis using LongDat 31 to identify bacterial taxa regulated over time post-KT. Post-transplantation, typical SCFA-producing genera like Coprococcus 34 , Lachnospiraceae 35 , Roseburia 36 , Faecalibacteria 37 and Ruminocococcus torques group 38 increased significantly over time (Fig 1E), suggesting an improvement of CKD-associated microbiome alterations, specifically the impaired production of SCFA as one of its hallmarks 6,8 .…”

“…For the longitudinal analysis, LongDat (v.1.1.2) 31 was used, building negative binomial generalized linear mixed models over time after KT, controlling for sample origin by a random intercept. The models were built on rarefied data with default filtering 31 . FDR was controlled by Benjamini-Hochberg procedure.…”

Airway microbiome signature accurately discriminates Mycobacterium tuberculosis infection status

“…Furthermore, the sensitivity and specificity of microbiome signatures must be carefully assessed, considering several factors such as prior antibiotic use, different geographic settings, and potential confounders such as age, sex, body mass index, nutrition, HIV status, respiratory comorbidities, and smoking status. However, with robust biostatistical tools that control for microbiome confounders in microbiome-wide association studies such as MetadeconfoundR, 68 , 69 and LongDat, 70 such hurdles can easily be overcome. In our study, a careful interrogation of potential confounders such as COPD, HIV, and antibiotic use was carefully considered.…”

MIAI: Maturation of Immunity Against Influenza - a German birth cohort study with focus on the development of immunity against respiratory viral infections

“…To elucidate the relationships between specific immunotypes, respiratory or gut microbiome profiles, environmental factors, and clinical outcomes, we have created two specialized software tools, metadeconfoundR (48) and longdat (49). These tools enable comprehensive mixed omics data analysis while managing the complex interplay of numerous demographic, therapeutic, and disease-related variables.…”