Differential transcript usage in the Parkinson's disease brain

This repository holds all code and data to reproduce results, figures and tables presented in "Differential transcript usage in the Parkinson's disease brain". Direct access to result statistics, raw counts (in TPM) and sample metadata (covariates used in the analysis) are available at: Figshare. The main analysis was performed with DEXSeq, DRIMSeq and stageR. The application of these tools for DTU analysis was performed as suggested in Swimming downstream: statistical analysis of differential transcript usage following Salmon quantification..

Prerequisites

• R >= 3.6.0
• Rscript in "$PATH"
• Bash
• All dependencies listed in DESCRIPTION of the DTU R package

How to..

Use uploaded result objects (.rds) to generate figures and tables:

In your terminal execute:

git clone https://github.com/fifdick/DTU_in_PD_brain.git
cd DTU_in_PD_brain
tar -zxvf ./results/rds/obj.tar.gz
tar -zxvf ./results/external/dge_results/DESeqOut_CT.tar.gz
# R CMD check DTU 
R CMD build DTU
R CMD INSTALL DTU_1.0.tar.gz

To reduce the results of the gene set enrichment analysis with stringDB, we used the code in ./PathCluster.

# R CMD check PathCluster
R CMD build PathCluster
R CMD INSTALL PathCluster_0.2.tar.gz

In your R editor open makeFigures.Rmd and follow the instructions there.

Rerun the analysis from scratch:

git clone https://github.com/fifdick/DTU_in_PD_brain.git
cd DTU_in_PD_brain
# unpack the reference .gtf file 
tar -xzf ./referenceData/extract_me_gtf.gz
# install the R package to your R
# R CMD check DTU 
R CMD build DTU
R CMD INSTALL DTU_1.0.tar.gz
# make the pipeline executable
chmod +x runDTU.sh
# potentially change parameters according to the instructions in runDTU.sh
# make sure Rscript is in your path as it will be called by the bash script
./runDTU.sh

• The analysis is complete when "Successfully intersected gene lists" appears in stdout. Results are in ./results/rds/, direct tool results are in the respective folders in ./results/.

• Two more steps: extract the DGE results such that they can be loaded and install PathCluster. (Functions used to cluster the results of the gene set enrichment analysis; not necessary if you want to skip that step in makeFigures.Rmd.)

# extract DGE results, needed to compare and make figures and tables
tar -zxvf ./results/external/dge_results/DESeqOut_CT.tar.gz

# R CMD check PathCluster
R CMD build PathCluster
R CMD INSTALL PathCluster_0.2.tar.gz

• Open ./makeFigures.Rmd in your R editor and follow the instructions there.

Content

• Transcript quantifications obtained with Salmon for all samples in the study are in ./rawData/. Where each subdirectory holds the estimates for the sample identified with the subdir name.
• For annotation of the transcripts during DTU analysis with DEXSeq and DRIMSeq (when applying tximport::tximport for the tx2gene parameter), we used the reference file provided in ./referenceData. (Zipped to be able to fit on github.)
• Meta data including all covariates used in the model design are supplied in the .csv in ./metaData. The sample IDs match the subdirectory names in ./rawData/. The cell type estimations listed as covariates ("Microglia_Genes" and "Oligo_Genes") have been estimated as described in the paper. The code for the calculation can be found on: gitApp. The estimation has be performed on the replication and discovery cohort separately, and all samples of each cohort where included.
• The log of final run of the pipeline can be found in ./logs.
• The R object that holds all results is compressed in ./results/rds. It consists of all data necessary to generate tables and figures presented in the paper (except the qPCR wetlab and DGE results). It was generated by running the script ./runDTU.sh. All arguments and parameters that were used to generate these results can be found in the log file ./log/04-05-20.txt. We also need the result dataframe from the DTU analysis we ran without the covariates Microglia_Genes and Oligo_Genes. We provide only the dataframe as opposed to the whole result object, due to storage limitations. However, this can be generated by running ./runDTU.sh with the parameter batch="rin:sex:age_years" instead of batch="rin:sex:age_years:Microglia_genes:Oligo_Genes".
• The directory ./DTU contains all helper functions collected into an R package. These are used in both the general pipeline (runDTU.sh) and in the .Rmd file that generates tables and figures.
• The directory ./scripts holds Rscripts which are called by ./runDTU.sh to perferm the general DTU analysis. Basically just wrapper functions around DRIMSeq and DEXSeq functions. The results of these are then assembled by ./scripts/bindToolRes.R to generate the results found in ./results/rds. These hold all R objects needed to create figures and tables in ./makeFigures.Rmd.
• Differential gene expression results from "Common gene expression signatures in Parkinson’s disease are driven by changes in cell composition" which are used to compare DTU results to DGE results are in ./results/external/dge_results/
• qPCR results are summerised in a .csv in ./results/external/qPCR/
• Code used to generate plots and tables is in ./makeFigures.Rmd.

Built with

Bash

5.0.3(1)-release

R sessionInfo() output

> sessionInfo()
R version 3.6.0 (2019-04-26)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Pop!_OS 19.04

Matrix products: default
BLAS:   /usr/lib/x86_64-linux-gnu/blas/libblas.so.3.8.0
LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.8.
0

locale:
 [1] LC_CTYPE=en_GB.UTF-8       LC_NUMERIC=C
 [3] LC_TIME=en_GB.UTF-8        LC_COLLATE=en_GB.UTF-8
 [5] LC_MONETARY=en_GB.UTF-8    LC_MESSAGES=en_GB.UTF-8
 [7] LC_PAPER=en_GB.UTF-8       LC_NAME=C
 [9] LC_ADDRESS=C               LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_GB.UTF-8 LC_IDENTIFICATION=C

attached base packages:
 [1] stats4    parallel  grid      stats     graphics
 [6] grDevices utils     datasets  methods   base

other attached packages:
 [1] wiggleplotr_1.8.0
 [2] DEXSeq_1.30.0
 [3] RColorBrewer_1.1-2
 [4] AnnotationDbi_1.46.0
 [5] DESeq2_1.24.0
 [6] SummarizedExperiment_1.14.0
 [7] DelayedArray_0.10.0
 [8] GenomicRanges_1.36.0
 [9] GenomeInfoDb_1.20.0
[10] IRanges_2.18.1
[11] S4Vectors_0.22.0
[12] Biobase_2.44.0
[13] BiocGenerics_0.30.0
[14] BiocParallel_1.18.0
[15] DRIMSeq_1.12.0
[16] reshape2_1.4.3
[17] kableExtra_1.1.0
[18] gridExtra_2.3
[19] aggregation_1.0.1
[20] ggridges_0.5.1
[21] viridis_0.5.1
[22] viridisLite_0.3.0
[23] dplyr_0.8.3
[24] tidyr_1.0.2
[25] readr_1.3.1
[26] Unicode_12.0.0-1
[27] knitr_1.23
[28] STRINGdb_1.24.0
[29] DTU_1.0
[30] ggrepel_0.8.1
[31] matrixStats_0.54.0
[32] cowplot_1.0.0
[33] VennDiagram_1.6.20
[34] futile.logger_1.4.3
[35] ggplotify_0.0.4
[36] patchwork_1.0.0.9000
[37] ggpubr_0.2.1
[38] magrittr_1.5
[39] ggplot2_3.2.1
[40] nvimcom_0.9-83

loaded via a namespace (and not attached):
  [1] backports_1.1.5        Hmisc_4.2-0
  [3] plyr_1.8.4             igraph_1.2.4.1
  [5] lazyeval_0.2.2         splines_3.6.0
  [7] digest_0.6.22          htmltools_0.4.0
  [9] gdata_2.18.0           checkmate_1.9.4
 [11] memoise_1.1.0          cluster_2.0.7-1
 [13] limma_3.40.4           Biostrings_2.52.0
 [15] annotate_1.62.0        prettyunits_1.0.2
 [17] colorspace_1.4-1       blob_1.2.0
 [19] rvest_0.3.4            xfun_0.8
 [21] crayon_1.3.4           RCurl_1.95-4.12
 [23] genefilter_1.66.0      survival_2.43-3
 [25] glue_1.3.1             hash_2.2.6.1
 [27] gtable_0.3.0           zlibbioc_1.30.0
 [29] XVector_0.24.0         webshot_0.5.1
 [31] scales_1.0.0           futile.options_1.0.1
 [33] DBI_1.0.0              edgeR_3.26.5
 [35] Rcpp_1.0.3             plotrix_3.7-6
 [37] xtable_1.8-4           progress_1.2.2
 [39] htmlTable_1.13.1       gridGraphics_0.5-0
 [41] foreign_0.8-71         bit_1.1-14
 [43] Formula_1.2-3          sqldf_0.4-11
 [45] htmlwidgets_1.5.1      httr_1.4.0
 [47] gplots_3.0.1.1         acepack_1.4.1
 [49] pkgconfig_2.0.3        XML_3.98-1.20
 [51] nnet_7.3-12            locfit_1.5-9.1
 [53] tidyselect_1.0.0       rlang_0.4.6
 [55] munsell_0.5.0          tools_3.6.0
 [57] gsubfn_0.7             RSQLite_2.1.2
 [59] evaluate_0.14          stringr_1.4.0
 [61] bit64_0.9-7            caTools_1.17.1.2
 [63] purrr_0.3.2            formatR_1.7
 [65] xml2_1.2.0             biomaRt_2.40.3
 [67] compiler_3.6.0         rstudioapi_0.10
 [69] png_0.1-7              ggsignif_0.5.0
 [71] tibble_2.1.3           statmod_1.4.32
 [73] geneplotter_1.62.0     stringi_1.4.3
 [75] lattice_0.20-38        Matrix_1.2-15
 [77] vctrs_0.2.4            pillar_1.4.2
 [79] lifecycle_0.1.0        BiocManager_1.30.4
 [81] Tmisc_0.1.22           data.table_1.12.2
 [83] bitops_1.0-6           R6_2.4.1
 [85] latticeExtra_0.6-28    hwriter_1.3.2
 [87] KernSmooth_2.23-15     lambda.r_1.2.3
 [89] gtools_3.8.1           assertthat_0.2.1
 [91] chron_2.3-53           proto_1.0.0
 [93] withr_2.1.2            Rsamtools_2.0.0
 [95] GenomeInfoDbData_1.2.1 hms_0.5.0
 [97] rpart_4.1-13           rmarkdown_1.14
 [99] rvcheck_0.1.8          base64enc_0.1-3