Experimental Design
The purpose of specifying the experimental design is to provide MetaMorpheus with extra information about the spectra files you want to search, specifically each file's condition (e.g., normal or cancer tissue), the biological replicate, the fraction (if your samples are fractionated), and the technical replicate. This information is currently used to normalize intensities. In the future, we plan to extend this functionality to other aspects of MetaMorpheus (e.g., identification analysis). A Wiki page about normalization in FlashLFQ/MetaMorpheus is forthcoming.
The experimental design can be specified by clicking the "Experimental Design" button in the bottom-left part of the interface, under the mass spectra files. The experimental design window looks like this:
Requirements for each designation for experimental design are listed here:
- Condition - Cannot be blank
- Biorep - an integer, at least 1. Each condition must have continuous biorep numbers starting at 1. For example, bioreps 1, 3, and 4 are not valid because biorep 2 is missing. In this case you would label the bioreps as 1, 2, and 3.
- Fraction - an integer, at least 1. Each biorep must have continuous fraction numbers starting at 1. If your data is not fractionated, just enter 1 for all fractions. It is OK for two bioreps to have different total numbers of fractions. It is NOT recommended to use a biorep if it is missing a fraction with significant peptide intensity (e.g., if biorep 2 is missing fraction #5 out of 10 total fractions).
- Techrep - an integer, at least 1. Each fraction must have continuous techrep numbers starting at 1.
- If the experimental design is not defined, each protein will have one column of intensities for each individual spectra file.
- If the experimental design is defined, each protein will have one column of intensities for each biological replicate, regardless of how many fractions or technical replicates there are. Non-zero intensities are averaged across technical replicates. The fraction with the highest intensity is used as a peptide's quantity in that biological replicate, and protein intensities are calculated from the peptide intensities. If normalization is enabled, normalization happens before biorep-level intensity summarization.
The experimental design window saves a file called ExperimentalDesign.tsv to the folder with the first spectra file marked "Use". There is no practical reason why you would want to alter this file manually if you are using the GUI (graphical user interface). If you are using the command-line version of MetaMorpheus, however, you will need to create the .tsv file yourself manually. An example file is attached on the bottom of this page. Be sure to follow the "Formatting" rules.
Currently, the experimental design settings are only used for intensity (i.e., quantification) normalization. In the future, we plan to use these experimental design settings for PSM analysis.
Note: You do not need to download this if you're using the graphical version of MetaMorpheus. You only need to download this and edit it manually if you're running the command-line version of MetaMorpheus.