Correcting Farm effects
Dear all,
My experiment was performed in 2 different farms and I have 16S rRNA sequences from these 2 farms. When I analysed the data, I observed strong farm effects. How to correct the farm effects to identify the true treatment effect? Is it Ok to correct for the farm effect or need to analyse these 2 farms separately? I tried the combat function and got negative values in the feature table. I do not think this is Ok. Can anyone help me on this?
Thank you
• 805 views
•
link
updated
by
Ram
•
written
by
Jeyamalar •
•
written
by
Jeyamalar •
0 answers
No answers yet.
Log in to answer this question.
More posts like this
-
multiple condition and time course in RNA-seq
written by tiancaigg •
-
Batch correction confounded by outcome distribution across batches
written by boris •
-
ComBat before or after filtering
written by averagelife •
-
Strategies for merging RNAseq datasets
written by Nico80 •
-
Should batch correction be performed before or after feature selection in scRNA-seq
written by Lucy •
-
Best way to merge RNA-seq data from two sequencing runs of the same samples
written by unsupervised_learner •
-
Analysis of ratios (fold changes) from different microarray experiments with common control
written by matt.arno •
-
Correct for batch-effects which are not visible in pca-plot of raw-data (RNA-Seq)
written by Tobias.Wohland
-
Setup for multiple strains, multiple conditions, and replicates
written by montezuma •
-
News: Science Article On Social Influence Bias On Voting Patterns ("Herding") With Impact On Biosta…
written by Michael Dondrup
Add
farmas a covariate into your differential abundance model to correct for it.It would be helpful to include additional information about your study design and a description of the primary objectives of your study. As you probably know, farms can have very different environments and management practices and this can have a major impact on the microbiome. Therefore, it seems plausible that the effect of your independent variable may vary between farms. Given this, I would recommend testing for an interaction between your independent variable and farm, and then stratifying your analysis by farm in the presence of a statistically significant interaction.