In quantitative Omics research, the goal is to understand which analytes (gene, protein or metabolite) are perturbed between experimental conditions; therefore we carefully design our studies to explore these questions. The algorithms used within the Assembler application require this study design information to correctly compute the differences. Typically you will have 3 main classifications for the samples that you are analyzing in an omics study. At the top level, you have Experimental groups which will represent the various conditions that you are studying (ie. Normal, early stage disease, late stage disease…). Within each experiment group, you will likely have biological replicates of each condition; multiple individuals within a single experimental group (Biological Group). Finally, you may also perform technical replicates for all or some of the samples, these could be LC-MS replicates or sample preparation replicates (Technical group). As you assign the metadata to each datafile, every sample will have a unique combination of the 3 group levels.
The requirement for replicates is tied in with the functionality of the algorithms within the Assembler application. The data is first normalized and reproducibility metrics are computed which involves having replicates within each experimental condition. It is recommended that you have at least 3 replicates within each experimental condition such that reasonable statistics can be generated. Biological replicates are typically desired for optimal study design and for this algorithm having 3 or more is recommended, eliminating the need to use technical replicates. However, if you only have 2 biological replicates, then it is recommended that you perform some technical replicates such that you have more replicates within each experimental condition. There is no limit on how many replicates you can have within an experimental group.