ideas:单细胞个体水平差异表达分析

# ideas methods on Autism data ### step1a_dca_prepare_data.R Prepare input data for DCA. Only needed if want to try dca_direct in step1d. ### step1a_dca.sh Command line for running DCA. Only needed if want to try dca_direct in step1d. ### step1a_split_dca_outputs.R Splitting the outputs of dca according to difference cell types, to prepare part of inputs for step1d. Only needed if want to try dca_direct in step1d. ### step1b_DESeq2.R DESeq2 without covariates and with covariates. ### step1c_ideas.R Run MiRKAT and permanovas on distance matrix calculated based on nb(negative binominal) approach. ### step1d_dca_direct.R Run MiRKAT and permanovas on distance matrix calculated based on dca_direct(based on outputs from DCA) approach. Note that here when using `fit_method = "dca_direct"` in function `ideas_dist`, the variable `"rd"` in `var_per_cell` is actually not involved in the distance matrix computing, unlike when using `fit_method = "nb"`. This input item is included here only to keep the code format consistent for different `fit_method` options. ### step1e_combine_results.R Combine p value results from step1b (results only from approach with covariates), step1c and step1e. ### step1f_gsea.R Gene set enrichment analysis of results from step1e. ### step1l_qvalues.R Compute corresponding q values. ### cell_types.txt List all the 17 cell types for looping through using .sh files when needed.