ASAP-PD5D-MTG-snRNA-Workflow: ASAP-PD5D-MTG-snRNA-Workflow

Code applied to snRNA-seq data from the middle temporal gyrus, part of the ASAP PD5D Parkinson's Cell Atlas project. Analysis steps are split into scripts that should be run in numerical order. The workflow is designed to be used with outputs from the 10X Cell Ranger pipeline cellranger count. An aggregated matrix of all samples is required for the sample quality control script (snRNA-seq_Workflow_1_sample_QC_1.R). snRNA-seq_Workflow_2_cell_QC_for_count_matrix.R performs cell QC for each sample individually, which can be submitted for all samples by the script Batch_run_snRNA-seq_Workflow_2_cell_QC_for_count_matrix.pl, providing the cellranger count output matrix folder for every sample is in the run directory. After QC the sample matrices are then aggregated by snRNA-seq_Workflow_3_merge_count_matrices.pl. snRNA-seq_Workflow_4_GLMPCA_and_harmony.R then takes the aggregate matrix as an input and outputs a Seurat Object. snRNA-seq_Workflow_4_GLMPCA_and_harmony.R also reads in a set of bespoke metadata so this section will require editing by other users to account for their own set of metadata. Again, it should be noted that later code may depend upon specific metadata columns in the Seurat Object, and so will need to be manually adjusted by other users. For small datasets it may be easier to skip scripts 2 and 3 and do the cell quality control and data aggregation in R itself as part of a Seurat workflow by customizing script 4. The proceeding scripts (5-7) then all take the Seurat Object from the preceding script as their input.