Estimation of alternative splicing intensity at single cell levels

 
Description:
 Alternative splicing (AS) significantly contributes to cellular transcriptomic heterogeneity, and single-cell RNA-Seq is commonly employed to delineate this heterogeneity. However, due to high dropout rates, inevitable noise, and limited coverage, accurately characterizing splicing changes at single-cell level remains challenging. To address this, we present a computational framework called SCSES (Single-Cell Splicing EStimation) to improve the AS profiles, which imputes the junction count matrix and fills in missing AS alternations by sharing information across similar cells or events with data diffusion.
 
Instructions:
 Gene TPM file: This file is a gene expression matrix where rows represent genes, columns represent cells, and values are TPM.
 Splicing Event File: Each line in this file is a splicing event name.
 Junction Read Counts file:: This file is a junction read counts matrix where rows represent junctions, columns represent cells, and values are read counts.
 Task Name: Task Name
 Thread Number: Thread Number
 Species: Species, with options for "Human" or "Mouse"
 Genome Version: The version of reference genome, with options: "hg19", "hg38", or "mm10".
 Minimal Cell Count Containing an Event: The minimum number of cells that are required to contain an event, default = 10
 Minimal Read Count Supporting a Junction: The minimum number of reads that are required to support a junction, default = 5
 Maximal Dropout Ratio of Genes Expression: The maximal percentage of cells that are required to have at least 1 reads for a gene, default = 0.9
 Maximal Dropout Ratio of Splicing Events: The maximal percentage of cells that are required to have dropout for an event, default = 0.9
 Filter Out Mitochondrial Genes: If filter out mitochondrial genes or not, default = "true"
 Filter Out Ribosomal Genes: If filter out ribosomal genes or not, default = "true"
 Cell Similarity Feature: The features used to calculate cell similarity, with options: " EXP_RBP ", "RC", or "PSI", “EXP_RBP” by default.
 The Number of High Variable Cell Similarity Features: The number of high variable cell similarity features for PCA, default = 1000
 Min K cell: Minimum number of dynamic cell neighbors, default = 5
 Max K cell	: Maximum number of dynamic cell neighbors, default = 50
 Restart Probability of Random Walk: The restart probability of random walk, default = 0.2
 K Neighbor Event: The number of event neighbors, default = 10
 Convergence Thresholds of Random Walk: The convergence thresholds of random walk, default = 0.05