Pre-gating in FlowJo

[AEsch]

Hi SPADE folks,

I browsed the other issues but didn't find anything related to my specific problem -- apologies if this has already been addressed (or if it's a ridiculous question). My CyTOF-generated files work perfectly with the SPADE plugin in Cytoscape; however, if I gate out the dead cells and doublets in FlowJo 10.0.4 first and then try to analyze the file, I get an "execution halted" message. This happens even if I drop the file into FlowJo and then immediately export as an FCS3 file without gating anything, so I'm guessing there's some FlowJo parameter that needs adjusting. Before opening CyTOF files in FlowJo, I change the parameters to the CyTOF default (Preferences > Tools > Cytometers > CyTOF) and then open a new workspace. I've referred to Mike Leipold's help sheet to tinker with some of the settings, but unfortunately nothing seems to work. I've tried adjusting SPADE parameters too (target number of clusters, downsampling percentage), but it's always the same: the FlowJo-tainted file gives an error message, while the original CyTOF file works beautifully.

Here's an example of the script generated by the SPADE plugin:

BEGIN AUTOGENERATED DATA

FILE_TO_PROCESS="." CLUSTERING_MARKERS=c("(Pr141)Dd","(Tm169)Dd") PANELS=list(

• list(
• panel_files=c("flowjomcf7.fcs"),
• median_cols=NULL,
• reference_files=NULL,
• fold_cols=NULL
• )) TRANSFORMS=flowCore::arcsinhTransform(a=0, b=0.2) Scalable Robust Estimators with High Breakdown Point (version 1.3-02)

DOWNSAMPLED_EVENTS=NULL DOWNSAMPLING_TARGET_PCTILE=0.05 TARGET_CLUSTERS=50 CLUSTERING_SAMPLES=50000 DOWNSAMPLING_EXCLUDE_PCTILE=0.01 NODE_SIZE_SCALE_FACTOR=1.2 NORMALIZE="global" OUTPUT_DIR="output/"

END AUTOGENERATED DATA

LIBRARY_PATH=NULL NUM_THREADS <- 1 for (e in commandArgs()) {

• ta <- strsplit(e,"=",fixed=TRUE)
• if( ta[[1]][1] == "-num_threads") {
• NUM_THREADS <- ta[[1]][2]
• }
• if( ta[[1]][1] == "-file_to_process") {
• FILE_TO_PROCESS <- ta[[1]][2]
• }
• } Sys.setenv("OMP_NUM_THREADS"=NUM_THREADS) library("spade",lib.loc=LIBRARY_PATH) Loading required package: igraph0 LAYOUT_FUNCTION=layout.kamada.kawai SPADE.driver(FILE_TO_PROCESS, file_pattern="*.fcs", out_dir=OUTPUT_DIR, cluster_cols=CLUSTERING_MARKERS, panels=PANELS, transforms=TRANSFORMS, layout=LAYOUT_FUNCTION, downsampling_samples=DOWNSAMPLED_EVENTS, downsampling_target_pctile=DOWNSAMPLING_TARGET_PCTILE, downsampling_exclude_pctile=DOWNSAMPLING_EXCLUDE_PCTILE, k=TARGET_CLUSTERS, clustering_samples=CLUSTERING_SAMPLES) Downsampling file: ./flowjomcf7.fcs Execution halted

I don't have any experience with R, so these scripts are pretty much indecipherable to me. I've been beating my head against the wall with this issue for a while, so any help you can give will be very much appreciated. Thanks!

• Amanda

[zbjornson]

Apologies for no one responding. Are you still hoping to use SPADE? If so, please reopen this issue.

[AEsch]

Hey Zach,

No problem — it was a FlowJo issue, not SPADE. Thanks!

• Amanda

Amanda M. Esch, Ph.D. Knight Cancer Institute Oregon Health & Science University 3181 SW Sam Jackson Park Road BRB 540 | L346JG Portland, OR 97239 phone 503.494.3012 | fax 503.494.3688 esch@ohsu.edumailto:esch@ohsu.edu

From: Zach Bjornson notifications@github.com<mailto:notifications@github.com> Reply-To: nolanlab/spade reply@reply.github.com<mailto:reply@reply.github.com> Date: Tuesday, September 24, 2013 1:45 PM To: nolanlab/spade spade@noreply.github.com<mailto:spade@noreply.github.com> Cc: esch esch@ohsu.edu<mailto:esch@ohsu.edu> Subject: Re: [spade] Pre-gating in FlowJo (#55)

Apologies for no one responding. Are you still hoping to use SPADE? If so, please reopen this issue.

— Reply to this email directly or view it on GitHubhttps://github.com/nolanlab/spade/issues/55#issuecomment-25040896.