Support to handle `.parquet` output from Vizgen #7190

alikhuseynov · 2023-04-19T18:40:41Z

Dear Seurat Team,
Here is the PR to handle parquet files (see #7080)
Example to make Seurat objects from Cellpose output of Vizgen data..
Let me know if additional edits are needed..
Thanks
A.

alikhuseynov · 2023-05-12T20:32:28Z

Thanks for the comment.
Do you mean several public Vizgen datasets or in general Cellpose output data?

z = 3L arg for z-plane is a default and was there before my PR. In order to be consistent, I added it to select a particular plane from Cellpose segmentation output.
I'm not aware that segmented cell boundaries differ through all 7 sections. If that's the case, then one should implement some additional preprocessing, but here also, which z to choose is an open question.

alikhuseynov · 2023-05-24T09:35:37Z

..just to confirm:
Vizgen is using just one z-plane, ie middle plane of the tissue for Cellpose segmentation.

R/preprocessing.R

AustinHartman · 2023-05-24T18:01:20Z

R/preprocessing.R

+    workers.total = 12,
+    DTthreads.pct = NULL,
+    coord.space = "micron", 
+    use.cellpose.out = TRUE


Do you think we can infer use.cellpose.out based on the presence of cell_boundaries.parquet files in data.dir? I want to make sure all of the older Vizgen outputs are still supported without requiring this parameter to be set.

That should be possible.
Basically it will read cellpose_micron_space.parquet or cellpose_mosaic_space.parquet , else the older Vizgen output, right?

Something like that, but my understanding was the cellpose outputs generated by Vizgen would are always named cell_boundaries.parquet based on their user guide: https://vizgen.com/wp-content/uploads/2022/12/91600001_MERSCOPE-Instrument-User-Guide_Rev-F.pdf (see page 45 and 46). Are your parquet output files generated by Vizgen, and if so, do you know what version of their software was used?

Our data was generated by Vizgen, then we got the link to download them. I will check with them again but it's probably the cell_boundaries.parquet and cell_by_gene.csv and cell_metadata.csv files would be the official one.

I will update the things we discussed here, plus some additional sanity checks are needed on cell_boundaries.parquet
see recent discussion #7080
thanks

Thanks @alikhuseynov! Vizgen has also agreed to review this PR and make any necessary changes so users can handle cellpose segmentations

Presence of a parquet file doesn't mean its a cellpose segmentation. Parquet files could be generated by either of the segmentation algorithms (see pg 29 of guide above).

Presence of a parquet file doesn't mean its a cellpose segmentation. Parquet files could be generated by either of the segmentation algorithms (see pg 29 of guide above).

yes, the idea is to allow users to load segmentation data (Cellpose or not) if .parquet is present. Some may have older data, ie .hdf5 files.

R/convenience.R

dfhannum · 2023-06-02T14:03:21Z

R/convenience.R

-  )
-  # add coords to seurat object
+LoadVizgen <- function(data.dir, fov = "vz", assay = 'Vizgen',
+                       add.zIndex = TRUE, update.object = TRUE,


It would useful to include the filter = '^Blank-' to filter out the blank genes.

Also, need to have mol.type = 'microns' as an input variable.

..could add optionallyfilter = '^Blank-', some users might want to keep the background genes for later checks.
..mol.type arg can be used in LoadVizgen()

I agree some users may want to look at them, but they should be removed before downstream analysis (and this way it works with the current Seurat Vizgen tutorial).

mol.type is also called outside of ReadVizgen() and it's producing errors for me when not loaded into the function.

sounds good, will add filter = "^Blank-" as default, if users want to keep them, then they should set filter = NA_character_ before loading data.

..will make sure that mol.type works without errors.

Working off your latest commit I added it mol.type = 'microns to LoadVizgen() arguments and it worked for all my test cases so far.

Great! For the next commit, I will make sure mol.type arg. work without specifying it in LoadVizgen()

..reads old-new data, optimized parallelization, few bugs fixed

..supports args: `mol.type`, `filter`, `z` added `verbose` for optional verbosity

alikhuseynov · 2023-06-06T10:35:34Z

The last commit should fix most of the bugs, supports old and new Vizgen data outputs, and optimized parallelization
If any bugs or something I missed, let me know please.

Test run would be this

## test =================================================
## make sure all libs are installed a priori!
# load libs ----
suppressPackageStartupMessages({
  library(ggplot2)
  library(Seurat)
  library(dplyr)
  library(magrittr)
  library(BiocParallel)
  library(progressr)
  library(spatstat)  
})

# helper function to return Seurat object metadata
callmeta <- function (object = NULL) { 
  return(object@meta.data) 
  }

# set directory to read from
dir_use <- "./merfish_seurat_debug/" # or something like "./region_0/"

##
start.time <- Sys.time()
obj <-
  LoadVizgen(data.dir = dir_use,  
             fov = "merfish.test", 
             assay = "Vizgen",
             metadata = c("volume", "fov"), # add cell volume info
             type = c("segmentations", "centroids"), # type of cell spatial coord matrices
             z = 3L,
             add.zIndex = TRUE, # add z slice section to a cell
             update.object = TRUE,
             use.BiocParallel = TRUE,
             workers.MulticoreParam = 14, # for `BiocParallel` processing
             verbose = TRUE
             )
end.time <- Sys.time()
message("Time taken to load object = ", 
        round(end.time - start.time, digits = 2), " ", 
        attr(c(end.time - start.time), "units"))
obj
obj %>% callmeta %>% str

sessionInfo()

R version 4.2.1 (2022-06-23)
Platform: x86_64-conda-linux-gnu (64-bit)
Running under: CentOS Linux 7 (Core)

Matrix products: default
BLAS/LAPACK: /xxx/envs/spatial_develop/lib/libopenblasp-r0.3.21.so

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

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] spatstat_2.3-4        spatstat.linnet_2.3-2 spatstat.core_2.4-4  
[4] rpart_4.1.16          nlme_3.1-159          spatstat.random_3.0-1
[7] spatstat.geom_3.0-3   spatstat.data_3.0-0   progressr_0.11.0     
[10] BiocParallel_1.30.4   magrittr_2.0.3        dplyr_1.1.2          
[13] SeuratObject_4.1.3    Seurat_4.3.0.9002     ggplot2_3.4.0        

loaded via a namespace (and not attached):
[1] readxl_1.4.1           uuid_1.1-0             backports_1.4.1       
[4] plyr_1.8.7             igraph_1.3.5           repr_1.1.4            
[7] lazyeval_0.2.2         sp_1.5-0               splines_4.2.1         
[10] listenv_0.8.0          scattermore_0.8        digest_0.6.29         
[13] htmltools_0.5.3        fansi_1.0.3            tensor_1.5            
[16] googlesheets4_1.0.1    cluster_2.1.4          ROCR_1.0-11           
[19] tzdb_0.3.0             readr_2.1.3            globals_0.16.1        
[22] modelr_0.1.9           matrixStats_0.62.0     spatstat.sparse_3.0-0 
[25] colorspace_2.0-3       rvest_1.0.3            ggrepel_0.9.2         
[28] haven_2.5.1            crayon_1.5.2           jsonlite_1.8.0        
[31] survival_3.4-0         zoo_1.8-10             glue_1.6.2            
[34] polyclip_1.10-0        gtable_0.3.1           gargle_1.2.0          
[37] leiden_0.4.2           future.apply_1.9.0     abind_1.4-5           
[40] scales_1.2.1           DBI_1.1.2              miniUI_0.1.1.1        
[43] Rcpp_1.0.9             viridisLite_0.4.1      xtable_1.8-4          
[46] reticulate_1.26        bit_4.0.4              htmlwidgets_1.5.4     
[49] httr_1.4.4             RColorBrewer_1.1-3     ellipsis_0.3.2        
[52] ica_1.0-3              pkgconfig_2.0.3        uwot_0.1.14           
[55] dbplyr_2.2.1           deldir_1.0-6           utf8_1.2.2            
[58] tidyselect_1.2.0       rlang_1.1.1            reshape2_1.4.4        
[61] later_1.3.0            munsell_0.5.0          cellranger_1.1.0      
[64] tools_4.2.1            cli_3.4.1              generics_0.1.3        
[67] broom_1.0.1            ggridges_0.5.4         evaluate_0.16         
[70] stringr_1.4.1          fastmap_1.1.0          goftest_1.2-3         
[73] bit64_4.0.5            fs_1.5.2               fitdistrplus_1.1-8    
[76] purrr_1.0.1            RANN_2.6.1             pbapply_1.7-0         
[79] future_1.28.0          mime_0.12              xml2_1.3.3            
[82] arrow_9.0.0            hdf5r_1.3.5            compiler_4.2.1        
[85] plotly_4.10.0          png_0.1-7              spatstat.utils_3.0-1  
[88] reprex_2.0.2           tibble_3.2.1           stringi_1.7.8         
[91] forcats_0.5.2          lattice_0.20-45        IRdisplay_1.1         
[94] Matrix_1.5-1           vctrs_0.6.2            furrr_0.3.1           
[97] pillar_1.9.0           lifecycle_1.0.3        lmtest_0.9-40         
[100] RcppAnnoy_0.0.19       data.table_1.14.6      cowplot_1.1.1         
[103] irlba_2.3.5            httpuv_1.6.5           patchwork_1.1.2       
[106] R6_2.5.1               promises_1.2.0.1       KernSmooth_2.23-20    
[109] gridExtra_2.3          parallelly_1.32.1      codetools_0.2-18      
[112] MASS_7.3-57            assertthat_0.2.1       withr_2.5.0           
[115] sctransform_0.3.5      hms_1.1.2              mgcv_1.8-40           
[118] parallel_4.2.1         sfarrow_0.4.1          grid_4.2.1            
[121] tidyverse_1.3.2        IRkernel_1.3           tidyr_1.2.0           
[124] googledrive_2.0.0      Cairo_1.6-0            Rtsne_0.16            
[127] pbdZMQ_0.3-7           spatstat.explore_3.0-5 lubridate_1.8.0       
[130] shiny_1.7.2            base64enc_0.1-3

dfhannum · 2023-06-06T21:21:52Z

This worked on all my test data and scenarios.

The only other suggestion I would have is making molecules an optional input (since the file can be quite large, and isn't always used during analysis). Otherwise this all looks great!

alikhuseynov · 2023-06-07T07:47:00Z

This worked on all my test data and scenarios.

The only other suggestion I would have is making molecules an optional input (since the file can be quite large, and isn't always used during analysis). Otherwise this all looks great!

Great, thanks!
I will add support for molecules an optional input

dfhannum · 2023-07-19T13:04:53Z

Any updates for this pull request?

alikhuseynov · 2023-07-19T14:33:54Z

Any updates for this pull request?

..sorry I was a bit busy, will be adding a new arg add.molecules = TRUE in LoadVizgen() to have molecules FOV optionally added to the object.
Afterwards, it kind of done on my side. I could try to solve small conflicts of the branch as well.
Any further suggestions?
Thanks

dfhannum · 2023-07-20T20:28:21Z

Any updates for this pull request?

..sorry I was a bit busy, will be adding a new arg add.molecules = TRUE in LoadVizgen() to have molecules FOV optionally added to the object. Afterwards, it kind of done on my side. I could try to solve small conflicts of the branch as well. Any further suggestions? Thanks

I think it looks great and looking forward to it's use

..resolve some small conflict for `Read10X_Image()`

..adding of molecules to the object is optional `add.molecules = TRUE`

alikhuseynov · 2023-07-26T09:15:12Z

I did last updates, tested loading functions and tried to resolve some conflicts in preprocessing.R
so far looks fine on my side, let me know if any additional changes are needed.
Thanks)

tangboyun · 2023-08-02T09:19:44Z

I did last updates, tested loading functions and tried to resolve some conflicts in preprocessing.R so far looks fine on my side, let me know if any additional changes are needed. Thanks)

Dear alikhuseynov,
I've tried your latest patch on the Vizgen '.parquet' data. When using the 'Crop' function like this:

vizgen.obj <- LoadVizgen(data.dir = '/path/to/input_dir', fov = 'fov',verbose=TRUE)
vizgen.obj <- SCTransform(vizgen.obj, assay = 'Vizgen', clip.range = c(-10, 10), )
vizgen.obj <- RunPCA(vizgen.obj, npcs = 30, features = rownames(vizgen.obj))
vizgen.obj <- RunUMAP(vizgen.obj, dims = 1:30)
vizgen.obj <- FindNeighbors(vizgen.obj, reduction = 'pca', dims = 1:30)
vizgen.obj <- FindClusters(vizgen.obj, resolution = 0.3)
vizgen.crop <- Crop(vizgen.obj[['fov']],x=c(1250,3750),y=c(8000,9000),coords='plot')

 I got the following error:

Error in h(simpleError(msg, call)) : 
  error in evaluating the argument 'y' in selecting a method for function 'Overlay': cannot derive coordinates from non-numeric matrix
Calls: Crop ... stopifnot -> Polygon -> coordinates -> coordinates -> .local

 How can I fix this?

alikhuseynov · 2023-08-02T12:25:50Z

I did last updates, tested loading functions and tried to resolve some conflicts in preprocessing.R so far looks fine on my side, let me know if any additional changes are needed. Thanks)

Dear alikhuseynov, I've tried your latest patch on the Vizgen '.parquet' data. When using the 'Crop' function like this:

vizgen.obj <- LoadVizgen(data.dir = '/path/to/input_dir', fov = 'fov',verbose=TRUE)
vizgen.obj <- SCTransform(vizgen.obj, assay = 'Vizgen', clip.range = c(-10, 10), )
vizgen.obj <- RunPCA(vizgen.obj, npcs = 30, features = rownames(vizgen.obj))
vizgen.obj <- RunUMAP(vizgen.obj, dims = 1:30)
vizgen.obj <- FindNeighbors(vizgen.obj, reduction = 'pca', dims = 1:30)
vizgen.obj <- FindClusters(vizgen.obj, resolution = 0.3)
vizgen.crop <- Crop(vizgen.obj[['fov']],x=c(1250,3750),y=c(8000,9000),coords='plot')

 I got the following error:

Error in h(simpleError(msg, call)) : 
  error in evaluating the argument 'y' in selecting a method for function 'Overlay': cannot derive coordinates from non-numeric matrix
Calls: Crop ... stopifnot -> Polygon -> coordinates -> coordinates -> .local

 How can I fix this?

Hi there, I haven't done any changes to Crop() function. It might be that your x and y limits are outside the range of the tissue coordinates. Also note that, x and y axes labels are swapped in the ImageDimPlot():

Alternatively, plot your data first with ImageDimPlot(), and try to switch x and y args (ie, y=c(1250,3750),x=c(8000,9000)) in your Crop() run and see if it works.

tangboyun · 2023-08-03T01:11:12Z

I did last updates, tested loading functions and tried to resolve some conflicts in preprocessing.R so far looks fine on my side, let me know if any additional changes are needed. Thanks)

Dear alikhuseynov, I've tried your latest patch on the Vizgen '.parquet' data. When using the 'Crop' function like this:
vizgen.obj <- LoadVizgen(data.dir = '/path/to/input_dir', fov = 'fov',verbose=TRUE)
vizgen.obj <- SCTransform(vizgen.obj, assay = 'Vizgen', clip.range = c(-10, 10), )
vizgen.obj <- RunPCA(vizgen.obj, npcs = 30, features = rownames(vizgen.obj))
vizgen.obj <- RunUMAP(vizgen.obj, dims = 1:30)
vizgen.obj <- FindNeighbors(vizgen.obj, reduction = 'pca', dims = 1:30)
vizgen.obj <- FindClusters(vizgen.obj, resolution = 0.3)
vizgen.crop <- Crop(vizgen.obj[['fov']],x=c(1250,3750),y=c(8000,9000),coords='plot')
 I got the following error:
Error in h(simpleError(msg, call)) : 
  error in evaluating the argument 'y' in selecting a method for function 'Overlay': cannot derive coordinates from non-numeric matrix
Calls: Crop ... stopifnot -> Polygon -> coordinates -> coordinates -> .local
 How can I fix this?
Hi there, I haven't done any changes to Crop() function. It might be that your x and y limits are outside the range of the tissue coordinates. Also note that, x and y axes labels are swapped in the ImageDimPlot():

ImageDimPlot() flipping axes #6110

x and y axes labels swapped in ImageDimPlot #6132

Alternatively, plot your data first with ImageDimPlot(), and try to switch x and y args (ie, y=c(1250,3750),x=c(8000,9000)) in your Crop() run and see if it works.

Thank you for your kind suggestion. I've tried to swap the x and y coordinates but still got the same error. I'll open a new issue for my problem.

tangboyun · 2023-08-04T07:29:35Z

Hi alikhuseynov,
I find that without options(Seurat.object.assay.version = "v5"), I will fail at the Crop function. With options(Seurat.object.assay.version = "v5"), I will fail at GetAssay:

> library(Seurat)
> options(Seurat.object.assay.version = "v5")
# The following 3 packages are required to make 'LoadVizgen' happy
> library(dplyr)
> library(magrittr)
> library(spatstat)

# 'verbose' has no default value
> vizgen.obj <- LoadVizgen(data.dir='/path/to/vizgen',fov='fov',verbose=TRUE)
> vizgen.obj <- SCTransform(vizgen.obj, clip.range = c(-10, 10), assay='Vizgen')
Error in GetAssay.Seurat(object = object, assay = assay) : 
  Vizgen is not an assay present in the given object. Available assays are: 
> GetAssay(vizgen.obj,assay='Vizgen')
Error in GetAssay.Seurat(vizgen.obj, assay = "Vizgen") : 
  Vizgen is not an assay present in the given object. Available assays are:
> vizgen.obj@assays$Vizgen
Assay (v5) data with 300 features for 65319 cells
First 10 features:
 Ccnd2, Th, Lhx2, Cdh4, Pdgfra, Sox9, Rnd2, Ube2c, Col1a1, Npas1 
Layers:
 counts

Maybe this pull request should be rebased on the 'seurat5' branch?

alikhuseynov · 2023-08-05T13:57:26Z

Hi alikhuseynov, I find that without options(Seurat.object.assay.version = "v5"), I will fail at the Crop function. With options(Seurat.object.assay.version = "v5"), I will fail at GetAssay:

> library(Seurat)
> options(Seurat.object.assay.version = "v5")
# The following 3 packages are required to make 'LoadVizgen' happy
> library(dplyr)
> library(magrittr)
> library(spatstat)

# 'verbose' has no default value
> vizgen.obj <- LoadVizgen(data.dir='/path/to/vizgen',fov='fov',verbose=TRUE)
> vizgen.obj <- SCTransform(vizgen.obj, clip.range = c(-10, 10), assay='Vizgen')
Error in GetAssay.Seurat(object = object, assay = assay) : 
  Vizgen is not an assay present in the given object. Available assays are: 
> GetAssay(vizgen.obj,assay='Vizgen')
Error in GetAssay.Seurat(vizgen.obj, assay = "Vizgen") : 
  Vizgen is not an assay present in the given object. Available assays are:
> vizgen.obj@assays$Vizgen
Assay (v5) data with 300 features for 65319 cells
First 10 features:
 Ccnd2, Th, Lhx2, Cdh4, Pdgfra, Sox9, Rnd2, Ube2c, Col1a1, Npas1 
Layers:
 counts

Maybe this pull request should be rebased on the 'seurat5' branch?

All PRs go to the "develop" branch of Seurat.
Seurat "v5" is bit different with new/updated features & functions, I could add this support to my forked branch "seurat5" as well soon.
If you are using this PR patch to load Vizgen data, make sure to use SeuratObject version 4.1.3 (can be installed form CRAN)

few small edits for cleaner code.

..inspired by [`SpatialFeatureExperiment`](https://github.com/pachterlab/SpatialFeatureExperiment) using`sf` package to work/filter segmentation geometries. Updates: - adding support to efficiently filter segmentation polygons (removing small artifacts etc..) - added helper function `.filter_polygons()` (a bit modified version from `SpatialFeatureExperiment:::.filter_polygons`) for filtering given `min.area` - minimal polygon area as a threshold - allow for multiple segmentation polygons per cell (..still using single z-plane)

..optimized parallelization `BiocParallel` for `.filter_polygons()`, ie using 2 workers less that total provided workers

alikhuseynov · 2023-08-28T14:54:35Z

Thanks @lambdamoses, I got inspired by SpatialFeatureExperiment where sf package is used to handle and filter segmentation geometries.

Updates for this PR:

adding support to efficiently filter segmentation polygons (removing small artifacts etc..)
- added helper function .filter_polygons() (a bit modified version from SpatialFeatureExperiment:::.filter_polygons) for filtering given min.area - minimal polygon area as a threshold, if set to NULL the max. area (ie, the largest polygon) will be used.
allow for multiple segmentation polygons per cell (..still using single z-plane for now)

How to use example -> make Seurat objects (essentially the same with a small update)

@dfhannum if you would like re-run it on some of your Vizgen data as before, to ensure the robustness, would be awesome.
Thanks!

swangvzg · 2024-01-05T16:24:21Z

Hi Seurat Team,

I am with Vizgen. I just tested alikhuseynov:feat/vizgen with Vizgen data. This update works well. please see this file: https://github.com/swangvzg/seurat/blob/master/notebooks/Seurat_PR.html

@AustinHartman @dfhannum @alikhuseynov @tangboyun

Thanks,
Shichen

alikhuseynov · 2024-01-05T17:02:07Z

Hi Seurat Team,

I am with Vizgen. I just tested alikhuseynov:feat/vizgen with Vizgen data. This update works well. please see this file: https://github.com/swangvzg/seurat/blob/master/notebooks/Seurat_PR.html

@AustinHartman @dfhannum @alikhuseynov @tangboyun

Thanks, Shichen

Thanks @swangvzg! I'm glad it works.. It is still on Seurat "4.3.0.9002" and it works with SeuratObject "5.0.1"
I would say it's up to develop team to integrate this PR into Seurat v5, I can make any additional changes that are needed/requested.

@swangvzg & @dfhannum, btw, we are developing Bioc. SpatialFeatureExperiment (SFE) and a converter between it and Seurat (under development). There is readVizgen in PR, if you and other Vizgen members are interested in testing it as well.

alikhuseynov · 2024-01-11T11:35:21Z

I can confirm that it work with Seurat v5 as well (R version 4.3.1), tested on alikhuseynov:vizgen_seurat5 (which is essentially the most recent develop branch with PR #8298)

molecules · 2024-02-14T21:31:13Z

R/convenience.R

@@ -20,11 +20,11 @@ NULL
 #' @rdname ReadAkoya
 #'
 LoadAkoya <- function(
-  filename,


I recommend removing all of the space-only changes so that reviewers can focus on changes that actually make a functional difference. Thank you so much for all of your work on this!

ok, I think most of the space-only changes are now removed. Thanks

alikhuseynov added 8 commits April 13, 2023 18:07

updates for Vizgen

5fe5b89

updates for Vizgen

aa32cc6

updates for ReadVizgen()

56d983c

updates for LoadVizgen

115e45d

fix for argument 'metadata'

eff89e9

param args for LoadVizgen

87d9303

fix args for ReadVizgen

089740a

add support for future.apply

547000e

alikhuseynov mentioned this pull request Apr 19, 2023

Processing vizgen data #7080

Open

alikhuseynov mentioned this pull request Apr 27, 2023

Seurat V5 on MERFISH data without hdf5r files #7215

Closed

Vizgen support single .parquet file

f73dbdb

AustinHartman self-requested a review May 24, 2023 17:48

AustinHartman reviewed May 24, 2023

View reviewed changes

R/preprocessing.R Outdated Show resolved Hide resolved

AustinHartman reviewed May 24, 2023

View reviewed changes

alikhuseynov added 2 commits May 31, 2023 22:26

major fix for .parquet segmentations

5ca1069

fix for LoadVizgen

9389ce9

dfhannum reviewed Jun 2, 2023

View reviewed changes

R/convenience.R Outdated Show resolved Hide resolved

dfhannum reviewed Jun 2, 2023

View reviewed changes

alikhuseynov added 2 commits June 6, 2023 11:22

major fix for ReadVizgen()

dc029ac

..reads old-new data, optimized parallelization, few bugs fixed

fix for LoadVizgen()

8f7153e

..supports args: `mol.type`, `filter`, `z` added `verbose` for optional verbosity

alikhuseynov changed the title ~~Support to handle Cellpose output .parquet from Vizgen~~ Support to handle .parquet output from Vizgen Jun 7, 2023

alikhuseynov mentioned this pull request Jun 28, 2023

ImageFeaturePlot can not show when data is huge #7505

Open

alikhuseynov mentioned this pull request Jul 20, 2023

A questio about LoadVizgen #7588

Open

alikhuseynov added 4 commits July 26, 2023 09:50

update ReadVizgen()

799323d

..resolve some small conflict for `Read10X_Image()`

update LoadVizgen()

9ea4458

..adding of molecules to the object is optional `add.molecules = TRUE`

resolving some conflicts in preprocessing.R

38fad2a

..update preprocessing.R from develop

bf15b6e

tangboyun mentioned this pull request Aug 3, 2023

'Crop' Function Failed on Vizgen Data #7635

Closed

cleaning ReadVizgen

f8461ca

few small edits for cleaner code.

alikhuseynov mentioned this pull request Aug 21, 2023

convert Seurat to SFE and back pachterlab/SpatialFeatureExperiment#3

Open

alikhuseynov added 5 commits August 22, 2023 14:17

small bug fix in ReadVizgen

a7be25a

..updated LoadVizgen()

f43fed0

adding .filter_polygons()

69ba89d

optimized parallelization for .filter_polygons()

038271f

..optimized parallelization `BiocParallel` for `.filter_polygons()`, ie using 2 workers less that total provided workers

dcollins15 force-pushed the develop branch from a87fd5f to 41d19a8 Compare November 28, 2023 20:43

molecules reviewed Feb 14, 2024

View reviewed changes

rm space-only changes in convenience.R

9db188a

alikhuseynov mentioned this pull request Mar 18, 2024

Issue with LoadVizgen #8597

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Support to handle `.parquet` output from Vizgen #7190

Support to handle `.parquet` output from Vizgen #7190

alikhuseynov commented Apr 19, 2023

alikhuseynov commented May 12, 2023

alikhuseynov commented May 24, 2023

AustinHartman May 24, 2023 •

edited

alikhuseynov May 24, 2023

AustinHartman May 24, 2023

alikhuseynov May 24, 2023

alikhuseynov May 25, 2023

AustinHartman May 25, 2023

dfhannum May 26, 2023

alikhuseynov May 28, 2023

dfhannum Jun 2, 2023

alikhuseynov Jun 2, 2023 •

edited

dfhannum Jun 2, 2023

alikhuseynov Jun 2, 2023

dfhannum Jun 2, 2023

alikhuseynov Jun 2, 2023

alikhuseynov commented Jun 6, 2023

dfhannum commented Jun 6, 2023

alikhuseynov commented Jun 7, 2023

dfhannum commented Jul 19, 2023

alikhuseynov commented Jul 19, 2023

dfhannum commented Jul 20, 2023

alikhuseynov commented Jul 26, 2023

tangboyun commented Aug 2, 2023

alikhuseynov commented Aug 2, 2023 •

edited

tangboyun commented Aug 3, 2023

tangboyun commented Aug 4, 2023

alikhuseynov commented Aug 5, 2023

alikhuseynov commented Aug 28, 2023

swangvzg commented Jan 5, 2024

alikhuseynov commented Jan 5, 2024

alikhuseynov commented Jan 11, 2024

molecules Feb 14, 2024

alikhuseynov Feb 15, 2024

Support to handle .parquet output from Vizgen #7190

Are you sure you want to change the base?

Support to handle .parquet output from Vizgen #7190

Conversation

alikhuseynov commented Apr 19, 2023

alikhuseynov commented May 12, 2023

alikhuseynov commented May 24, 2023

AustinHartman May 24, 2023 • edited

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

alikhuseynov Jun 2, 2023 • edited

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

alikhuseynov commented Jun 6, 2023

dfhannum commented Jun 6, 2023

alikhuseynov commented Jun 7, 2023

dfhannum commented Jul 19, 2023

alikhuseynov commented Jul 19, 2023

dfhannum commented Jul 20, 2023

alikhuseynov commented Jul 26, 2023

tangboyun commented Aug 2, 2023

alikhuseynov commented Aug 2, 2023 • edited

tangboyun commented Aug 3, 2023

tangboyun commented Aug 4, 2023

alikhuseynov commented Aug 5, 2023

alikhuseynov commented Aug 28, 2023

swangvzg commented Jan 5, 2024

alikhuseynov commented Jan 5, 2024

alikhuseynov commented Jan 11, 2024

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Support to handle `.parquet` output from Vizgen #7190

Support to handle `.parquet` output from Vizgen #7190

AustinHartman May 24, 2023 •

edited

alikhuseynov Jun 2, 2023 •

edited

alikhuseynov commented Aug 2, 2023 •

edited