Dimensionality reduction was done using Seurat package in R. First, we selected variable genes by using the FindVariableGenes function. To identify variable genes across a range of expression values, this function first computes the mean expression and dispersion [i.e., log(variance/mean)] per gene. Then, it bins genes according to their mean expression and calculates z scores for dispersion within each bin. Last, it then uses the dispersion-based z scores to select variable genes from the different bins. Only genes with mean expression between 0.0125 and 2 and dispersion >0.9 were considered (fig. S2B), resulting in 2399 variable genes. These variable genes were then used as input to linear dimensionality reduction based on principal component analysis (PCA), executed via the RunPCA function. This function identifies the PCs that account for the largest variability in the data, in a decreasing order. We estimated the significance of PCs by using the JackStraw function. This function repeatedly permutes a subset of the genes and calculates the PC scores of each gene to assess the likelihood that the PC score of a gene was obtained by chance. Accordingly, significant PCs are those enriched for genes with low P values. To create the t-SNE projections, the first 20 PCs (all of which were significant, P < 7.2 × 10−18) were used via RunTSNE function by the elbow method (fig. S2C). Perplexity of the t-SNE projection was set to 10, 30 (default), 50, 100, and 150, and showed similar shapes of projections. Thus, default perplexity was used.

Note: The content above has been extracted from a research article, so it may not display correctly.

Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.

We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.