# ——— REQUIRED SETTINGS ————–

# example config file with all possible settings

# internal short name, only visible in the URL # same as the output directory name # no special chars, no whitespace, please name = “sample”

# priority determines the order of the datasets # smallest comes first priority = 10

# tags are shown in the dataset browser # current tags: # smartseq2,10x tags = [“smartseq2”]

# human-readable name of this dataset shortLabel=”CellBrowser 100-genes demo”

# name of the expression matrix file, genes are rows exprMatrix=”exprMatrix.tsv.gz”

# “gencode-human”, “gencode-mouse” or “symbol” # For “symbol” you can specify which database to use to check # symbols or, for cbHub, how to map them to the genome. # ‘auto’ will automatically detect Ensembl human/mouse IDs # and translate to symbols geneIdType=”auto”

# name of the meta data table (“samplesheet). One sample per row. First row is name of sample. meta=”meta.tsv”

# we try to auto-detect the field type of fields in the meta data. # Sometimes, this doesn’t work, e.g. when your cluster ID is a numer # or your C1 chip ID is a number, but you don’t want them binned, you want # to treat as if they were categories enumFields = [“c1_cell_id”]

# tsv files with coordinates of every sample in format <sampleId, x, y> # first the name of the file, then a human readable description coords=[

{

“file”:”tsne.coords.tsv”, “flipY” : False, # R/Matplotlib files need to be flipped on the Y-axis “shortLabel”:”t-SNE on WGCNA”

}, {

“file”:”subset.coords.tsv”, “shortLabel”:”neural cells”, # you can overlay lines onto the cells, table has to have columns named x1, x2, y1, y2 “lineFile” : “lines.tsv”, # you can flip the y-axis of just the lines, relative to the points # This was necessary for a user when using the files produced by the URD pseudotime package #”lineFlipY” : True, # you can automatically switch on coloring on a meta data field whenever a layout is activated “colorOnMeta”:”neuralCluster”

},

]

# default field in the meta data table with the name of the cluster clusterField=”WGCNAcluster”

# default field in the meta data table used for the label of the clusters shown by default labelField=”WGCNAcluster”

# ——— OPTIONAL SETTINGS ————–

# The settings below are used to create filters on a Cell Browser instance (e.g. cells.ucsc.edu). # Note, all filter values can be a list (e.g. body_parts = [“brain”,”cortex”]) body_parts = [“”] # body_parts = [“heart”] organisms = [“”] # organisms = [“Human (H. sapiens)”, “Mouse (M. musculus)”] diseases = [“”] # diseases = [“Healthy”] projects = [“”] # projects = [“GTEx”, “Human Cell Atlas”, “hca”]

# genes that are highlighted in your paper can be pre-loaded and are shown as a clickable table on the left # this is optional but we highly recommend that you define at least 2-3 quick genes, it makes the browser a lot # more intuitive for users quickGenesFile = “quickGenes.csv”

# if you want to enforce some order of the values of your enums, e.g. your cluster annotation should be sorted # in a given order in the display, supply a text file with the values in the right order, one per line. # You can supply one text file per meta data field # enumOrder = { “WGCNAcluster” : “clusterorder.txt” }

# tsv files with marker gene lists for the clusters # format is (clusterName, geneSymbol, pValue, enrichment) + any additional fields or URLs you want to show markers=[

{“file”:”markers.tsv”, “shortLabel”:”Cluster-specific markers”}

]

# do not show this dataset on the dataset list. This can be used for pre-publication data. # visibility=”hide”

# optional: UCSC track hub with the BAM file reads and expression values # Alternatively, you can also provide a full link to a UCSC Genome Browser session here hubUrl=”http://cells.ucsc.edu/cortex-dev/hub/hub.txt”

# optional: table with <name><color> for any meta data values # color is a six-digit hexcode # name is a any value in the meta data table, e.g. cluster name. Canb be a .tsv or .csv file. colors=”colors.tsv”

# should the cluster labels be shown by default (default: true) showLabels=True

# the radius of the circles. If not specified, reasonable defaults will be used #radius = 5 # the alpha/transparency of the circles. If not specified, reasonable defaults will be used. #alpha = 0.3

# you need short names for your clusters, as there is little space on the plot # but cell types have complicated and long names # So you can provide a table with two columns: 1) short cluster name 2) long version # e.g. EC, endothelial cells # can be a .tsv or .csv file acronymFile = “acronyms.tsv”

# the unit of the values in the expression matrix # any string, shown on genome browser and violin y-Axis # typical values are: “read count/UMI”, “log of read count/UMI”, “TPM”, “log of TPM”, “CPM”, “FPKM”, “RPKM” unit = “TPM”

# format of the numbers in the matrix. # ‘auto’ works in 99% of the cases. Otherwise you can use ‘int’ for integers and ‘float’ for floating point numbers. # Use ‘forceInt’ if your matrix contains only integers but in a format like 3.123e10 # or the matrix has only integers expressed like 100.000, 200.000, 300.00, … matrixType=’auto’

# rarely needed: if your expression matrix does not contain genes, but something # else, like “lipids” or “plankton”, you can replace the word “gene” in the # user interface with another word # geneLabel = “Lipid”

# the default color palettes for this dataset. By default, we use Paul Tol’s # but you could use other ones, see the URL when you change the palette to see possible values # defQuantPal = “viridis” # defCatPal = “rainbow”

# you can optionally show little images for clusters on the tooltip. # For now, they have to be PNGs. # For now, you will have to copy these images to the source destination htdocs directory manually # right now, only brain-lipids/all-lipids is using this # clusterPngDir = “clusterImgs”

# Enter organs/organ parts here in a comma-separated list. Organs entered will show up as filter options for dataset list. # Really only useful if you have a cell browser with multiple datasets covering multiple organs (e.g. cells.ucsc.edu) # If no datasets have ‘body_parts setting, filter won’t be displayed. # body_parts=[“eye”,”cornea”]