Principal Component Analysis (PCA) of Proteomics Data

In this video, we perform a principal component analysis (PCA) to identify outlier genes & between and within sample variabilities in a typical proteomics dataset.
We prepare the data matrix of the expressions, normalize the expressions by extracting their sample means from them and dividing the result by each gene's standard deviation. After that, we perform the actual PCA.

We make a plot that shows the variance of each principal component (PC). The first principal component contains the largest variance of the dataset, the second principal component contains the second largest variance, etc.

Usually, the first two PC should explain most of the variation in the data. Here we see if that is the case

We usually plot the first principal component, containing the largest change, versus the second principal component, containing the second largest change. An ideal dataset has a PCA plot in which the samples under the same experimental conditions are very close to each other while samples under different experimental conditions are far away.
——
Document: https://compu-flair.com/pca
Code: https://colab.research.google.com/dri...

Видео Principal Component Analysis (PCA) of Proteomics Data канала CompuFlair