Hello,

I am working on a shotgun metagenomics data. Briefly, after contig assembly, the ORFs were predicted usinf Prodigal, and annotations were carried against different databases including KEGG Orthology (KO) via DRAM (Distilled and Refined Annotation of Metabolism) tool. For now, I am interested in KO-annotated genes, only, and trying to figure out an appropriate normalization method. My ultimate goal is to compare the abundances between sample groups, and annotate the pathways.

For each sample, I calculated the TPMs of each gene (despite the arguments against TPM for inter-sample calculations). Then, extracted the KO TPM abundances for each sample, and prepared the abundance matrix for all samples. I would like to ask your suggestions, whether additional normalization and standardization method is necessary, as it would drastically effect the downstream statistical analyses.

Thank you in advance.

Best,