Merge pull request #34 from rnabioco/manuscript

tidy up repo

README.Rmd

@@ -23,7 +23,7 @@ knitr::opts_chunk$set(
 
 Cell-type annotations are frequently excluded from public single cell datasets. This hinders single cell sequencing analysis reproducibility and accessibility. To better describe the issue, we monitor GEO entries monthly (currently set to auto-update at 1AM UTC, 1st of the month), and programmatically determine the fraction of entries with (potentially, likely overestimated) usable cell metadata. 
 
-Descriptions of the issue and suggestions (in short: PLEASE PLEASE deposit some metadata at cell level for scRNA-seq data) are in this short [manuscript](https://www.biorxiv.org/content/10.1101/2020.11.20.391920v1).
+Descriptions of the issue and suggestions (in short: PLEASE PLEASE deposit some metadata at cell level for scRNA-seq data) are in this short [manuscript](https://www.biorxiv.org/content/10.1101/2020.11.20.391920v1). Supplemental data files related to the manuscript are located in the `inst/manuscript/` directory.
 
 The latest archive of GEO scRNA-seq records with other associated data can be directly downloaded here [current_geo.rds](https://github.com/rnabioco/someta/raw/master/inst/extdata/current_geo.rds). GEO filtering, preview [first lines of text files, and content of .tar], and spot check feedback can be accessed from last tab on the `clustifyr` [web app](https://raysinensis.shinyapps.io/clustifyr-web-app/?tab=someta).
 

inst/extdata/workflow.gv

@@ -9,8 +9,8 @@ digraph workflow {
 
   # start
   node [fillcolor = "#E69F00"]
-  startnode [label = "Examine GEO record"];
-  mdata [label = "Load data into R/Python\nand reformat for reanalysis"]
+  startnode [label = "Examine data\n repository record"];
+  mdata [label = "Load data into R or Python\nand reformat for reanalysis"]
 
 
   # rds or h5ad parsing
@@ -35,7 +35,7 @@ digraph workflow {
   # post metadata processing
   node [fillcolor = "#E69F00"]
   post [label = "Generate reference dataset"]
-  post_2 [label = "Compare query single cell data\nto reference dataset"]
+  post_2 [label = "Compare query and reference\nsingle cell datasets\n (Annotate cell types, \nperform dataset integration, \ncompare gene signatures, etc.)"]
 
   # query data
   node [fillcolor = "#F0E442"]
@@ -46,8 +46,8 @@ digraph workflow {
         fontname = "Helvetica"]
 
   startnode -> mdata [
-                      label = "Has a UMI\ncount matrix?"]
-  startnode -> nomdata_nomatrix [label = "No UMI count matrix?"]
+                      label = "Has a \ncount matrix?"]
+  startnode -> nomdata_nomatrix [label = "No count matrix?"]
   nomdata_nomatrix -> mdata
 
   mdata -> nomdata_3 [xlabel = "No code\nprovided"]