10 Glossary
- PCA
- Principal Component Analysis (PCA) is a method of dimensionality reduction and to identify factors that drive the variation in a dataset (https://setosa.io/ev/principal-component-analysis)
- PLS-DA
- Partial Least-Squares Discriminant Analysis (PLS-DA) is a machine learning method that can be used for discriminant analysis or classification. It allows sharpening the difference between condition groups by projecting corresponding data variance onto hyperplanes (components), similar to the principal components of a PCA (https://en.wikipedia.org/wiki/Partial_least_squares_regression)
- Peptides
- Proteins can be digested by enzymes into peptides. In mass spectrometry based proteomics proteins are usually digested by trypsin during sample preparation which specifically cleaves at lysine and arginine residues.
- Protein Groups
- Proteins that share peptides are forming a protein group, e.g. if all detected peptideds of protein A also belong to protein B, the proteins A and B form one protein group.
- discriminant analysis
- A technique hat is used to analyze the data when the the dependent variable is categorical (e.g. “Healthy” vs “Disease”) and the independent variable is interval in nature. The objective of discriminant analysis is to develop discriminant functions that are nothing but the linear combination of independent variables that will discriminate between the categories of the dependent variable in a perfect manner.
- z-score standardization
- all values are standardized by x - mu / sd. The zscored values always follow a distribution with mean of 0 and a standard deviation of 1
- Protein Groups: Proteins that share peptides are forming a protein group, e.g. if all detected peptideds of protein A also belong to protein B, the proteins A and B form one protein group.↩︎
- Peptides: Proteins can be digested by enzymes into peptides. In mass spectrometry based proteomics proteins are usually digested by trypsin during sample preparation which specifically cleaves at lysine and arginine residues.↩︎
- z-score standardization: all values are standardized by x - mu / sd. The zscored values always follow a distribution with mean of 0 and a standard deviation of 1↩︎
- PCA: Principal Component Analysis (PCA) is a method of dimensionality reduction and to identify factors that drive the variation in a dataset (https://setosa.io/ev/principal-component-analysis)↩︎
- PLS-DA: Partial Least-Squares Discriminant Analysis (PLS-DA) is a machine learning method that can be used for discriminant analysis or classification. It allows sharpening the difference between condition groups by projecting corresponding data variance onto hyperplanes (components), similar to the principal components of a PCA (https://en.wikipedia.org/wiki/Partial_least_squares_regression)↩︎
- discriminant analysis: A technique hat is used to analyze the data when the the dependent variable is categorical (e.g. “Healthy” vs “Disease”) and the independent variable is interval in nature. The objective of discriminant analysis is to develop discriminant functions that are nothing but the linear combination of independent variables that will discriminate between the categories of the dependent variable in a perfect manner.↩︎