Figure - Omics Layers in Microbiology
Which layer answers which question — and why genome alone is not enough.
flowchart TB G["Genome<br/>(what it CAN do)"] --> T["Transcriptome<br/>(what it IS doing)"] T --> P["Proteome<br/>(what is actually made)"] P --> M["Metabolome<br/>(what it produces)"] P --> S["Structure<br/>(how it works)"] G --> Ph["Phenotype<br/>(what we observe)"] T --> Ph P --> Ph M --> Ph G -.->|"[[Microbial Genomics]]"| G2[WGS, assembly, annotation] T -.->|"[[Microbial Transcriptomics]]"| T2[RNA-seq, dual RNA-seq] P -.->|"[[Proteomics and MALDI Bioinformatics]]"| P2[MALDI-TOF, LC-MS/MS] S -.->|"[[Structural Bioinformatics]]"| S2[AlphaFold, cryo-EM, docking] Ph -.->|"ground truth"| Ph2[[Antimicrobial Susceptibility Testing]]
Key idea
Genotype-based prediction fails exactly where the causal step lives above the genome: efflux upregulation, porin loss, inducible enzymes, and persistence are transcript- or protein-level phenomena — see Genotype to Phenotype Prediction.