Hi All, any suggestions on using this pipeline for Visium HD data (multiple samples)? Thanks!

Hello developers and users of #C031QFFTT7G , I just wanted to forward this invitation to you, since I think your input and contribution could be very valuable:

1st iO-FAST (initiative for FAIR spatial data) think tank meeting

11 December 2024

TITANIC Hotels Berlin and online.

With the rapid evolution of spatial omics technologies, datasets are becoming increasingly complex, presenting challenges in data sharing, standardisation, and metadata annotation. The diversity of proprietary formats used in the field makes it difficult to achieve harmonised data analysis across labs. While sequencing data has made progress in standardisation, these efforts are still lacking on a broader scale for spatial data. The IO-FAST meeting aims to address these challenges by:

• Collaborating with academia, publishers, data repositories, and industry to streamline data representation.

• Providing guidance on metadata annotation and digital objects.

• Establishing synergies within the community to connect diverse efforts in data harmonisation.

• Offering workshops and training materials for PhDs, postdocs, and other researchers to promote data sharing policies.

We invite you to join this collaborative effort to shape the future of FAIR spatial data. This will be an excellent opportunity to share use cases, exchange ideas, and help develop frameworks for standardising spatial data formats and workflows. Please sign up to the meeting using the registration form below:https://indico.dkfz.de/event/1112/.

If you have any queries please contact Amonida Zadissa (aonida.zadissa@ukdri.ac.uk) or Jan-Phillip Mallm (j.mallm@Dkfz-Heidelberg.de).

On behalf of the IO-FAST organising committee:

Jan-Phillip Mallm, Head of Single Cell Open Lab, DKFZ

Sam Jackson, Tools and Technology Platform Manager, UK DRI

Amonida Zadissa, Associate Director of Informatics, UK DRI

Hi Spatial Vi Team, Thank you so much for developing such a useful pipeline! I have a question regarding slide file provisioning for Space Ranger. Since I don’t have internet access on my compute node, I need to manually provide the slide file I downloaded from the 10X Genomics website. I attempted to add it in my params file and as command line arguments using --

spaceranger_slidefile

and also tried

--slidefile

, but I haven’t been able to resolve the issue. Could you recommend the best way to accomplish this? Thank you very much for your time and help! Best regards, Thomas

Hi Spatial user, I try to use this pipeline by running a test run:

nextflow run nf-core/spatialvi \

-profile singularity,test *-r* dev

However, I got error related to render notebook quarto render I think. My I know the code that you run successful? Thank you so much!

Hi Spatial Team, I'm new with ST and nf-core. I'm slightly confused because .tif files are named by date, but there is no information about which samples are. Another thing is how to get the info corresponding to the slide and area. If someone can help me, please. 😇

Hi spatial users, Hope this question is still related. I try to annotate cell type from visium dataset of brain using a scRNA-seq reference. I apply the method used here: https://satijalab.org/seurat/articles/spatial_vignette. However, my result got zero score and I don’t know what went wrong. May I have your comment? Thank you!

Hey spatialvi team, I tried running yesterday with

test_full

on AWS and then realized the samplesheet for that test doesn't exist. What is the current full test that people run on the pipeline?

Hi spatialvi users, just curious if any of you can run successful the pipeline using singularity? I ran a simplest command but still get error message (discussed with Erik above).

nextflow run nf-core/spatialvi -profile singularity,test -r dev

Thank you.

Hey spatialvi team, thanks for the super useful pipeline. I was wondering if there is a plan to add cell2bin to the visium HD workflow? Would be super useful. Thanks for the info, much appreciated.

Hi folks - @David Lecomber and I are poking around in this pipeline trying to get linux/arm64 support running (see the #C07S9BW2STT channel). I was wondering what env/environment.yml file is used for? It struck me that it looks quite a lot like the shared

nf-core/quartonotebook

module (with its own environment file). Is it possible that the former is a historical relic and not used anywhere? Or have I just missed where it is?

Separate side note - we should not have personal dockerhub accounts used in pipelines, as is done here. If you really need custom docker images, these should be published on the shared nf-core quay.io account, so that they can be maintained centrally rather than relying on a single individual.

Or better still, don't use a custom container at all and grab one from biocontainers or Seqera Containers..

Hi! Singlecell/spatial newbie here. How do you get to the fastq files? For the scrnaseq pipelines there is nf-core/demultiplex but AFAICS it does not support

spaceranger mkfastq

It would be great if one of the pipelines would support demultiplexing for spaceranger.

Or are there some settings for bcl2fastq in demultiplex that replicate what spaceranger mkfastq does?

Hi All, I was looking at the input file structure required by the pipeline, and I noticed that it starts from the folder where the FASTQ files are placed, rather than from a list of FASTQ files as in nfcore/single-cell pipelines. Do you plan to implement this feature in the future to allow direct reading of FASTQ files?

Hi, first of all thanks a lot for developing this! I was wondering today if I can run Probe Based CytAssist enabled samples on the pipeline, as it complained about the probeset csv missing. Best, Niko

Hello all! Thank you for such a helpful pipeline. I apologize if this has been asked before, but I am new to this channel. I am running several spatial experiments with spatialvi and they are all throwing the same error: "ERROR ~ Error executing process > 'NFCORE_SPATIALVISPATIALVIREAD_DATA (1) Caused by: Process

NFCORE_SPATIALVI:SPATIALVI:READ_DATA

input file name collision -- There are multiple input files for each of the following file names: sample name/scalefactors_json.json, sample name/raw_feature_bc_matrix.h5,sample name/tissue_hires_image.png, sample name/tissue_lowres_image.png" Does anyone have any suggestions for me to troubleshoot this?

Hi Raven, I just saw your Github issue, I will do some tests!

Hi Spatialvi team, Thank you for working on this pipeline! I'm trying to run it for the first time, and I'm getting an error in

spatially_variable_genes.qmd

:

File /opt/conda/lib/python3.10/site-packages/squidpy/gr/_nhood.py:185, in nhood_enrichment
  _test = _create_function(n_cls, parallel=numba_parallel)
  ...
RecursionError: maximum recursion depth exceeded

My data was processed successfully in earlier steps, so I'm wondering if the complexity is too high for this one, or the issue lies elsewhere. Have you come across this before? Any help would be much appreciated!

Hey spatialvi team, @Jonathan Manning and I are currently looking to run spatialvi on AWS and encountered some issues we fixed: https://github.com/nf-core/spatialvi/pull/122. In general, we can merge these PRs ourselves, but just wanted to give a quick heads up to@Erik Fasterius as the main developer?

Hej @Florian Wuennemann, thanks a lot, yes you can merge the PR, looks good to me!

Hi there, it's me again. I was running into the following error while trying to process some VISIUM HD data with the pipeline. Looks like the default settings can't necessarily detect images for VISIUM HD properly?

Command error:
  10:45PM INF instance metadata ami-id=ami-047eeab72a1800334 instance-id=i-01546edf255ede790 instance-type=r6id.4xlarge
  [WARN  tini (38)] Tini is not running as PID 1 and isn't registered as a child subreaper.
  Zombie processes will not be re-parented to Tini, so zombie reaping won't work.
  To fix the problem, use the -s option or set the environment variable TINI_SUBREAPER to register Tini as a child subreaper, or run Tini as PID 1.
  /opt/conda/lib/python3.10/site-packages/anndata/_core/anndata.py:1756: UserWarning: Variable names are not unique. To make them unique, call `.var_names_make_unique`.
    utils.warn_names_duplicates("var")
  /usr/local/bin/read_data.py:63: UserWarning: No full resolution image found. If incorrect, please specify the path in the `fullres_image_file` parameter when calling the `visium_hd` reader function.
    spatialdata = spatialdata_io.visium_hd(
  Traceback (most recent call last):
    File "/usr/local/bin/read_data.py", line 63, in <module>
      spatialdata = spatialdata_io.visium_hd(
    File "/opt/conda/lib/python3.10/site-packages/spatialdata_io/readers/visium_hd.py", line 308, in visium_hd
      load_image(
    File "/opt/conda/lib/python3.10/site-packages/spatialdata_io/readers/visium_hd.py", line 99, in load_image
      _load_image(
    File "/opt/conda/lib/python3.10/site-packages/spatialdata_io/readers/visium_hd.py", line 378, in _load_image
      data = imread(path, **imread_kwargs)
    File "/opt/conda/lib/python3.10/site-packages/dask_image/imread/__init__.py", line 48, in imread
      with pims.open(sfname) as imgs:
    File "/opt/conda/lib/python3.10/site-packages/pims/api.py", line 199, in open
      raise UnknownFormatError(
  pims.api.UnknownFormatError: All handlers returned exceptions:
  <class 'pims.image_reader.ImageReader'> errored: Could not find a backend to open `Visium_HD_Human_Colon_Normal_P5/binned_outputs/square_002um/spatial/tissue_hires_image.png`` with iomode `r`.
  Based on the extension, the following plugins might add capable backends:
    pyav:  pip install imageio[pyav]
    opencv:  pip install imageio[opencv]
  <class 'pims.image_reader.ImageReaderND'> errored: Could not find a backend to open `Visium_HD_Human_Colon_Normal_P5/binned_outputs/square_002um/spatial/tissue_hires_image.png`` with iomode `r`.
  Based on the extension, the following plugins might add capable backends:
    pyav:  pip install imageio[pyav]
    opencv:  pip install imageio[opencv]
  <class 'pims.imageio_reader.ImageIOReader'> errored: Could not find a backend to open `Visium_HD_Human_Colon_Normal_P5/binned_outputs/square_002um/spatial/tissue_hires_image.png`` with iomode `r?`.
  Based on the extension, the following plugins might add capable backends:
    pyav:  pip install imageio[pyav]
    opencv:  pip install imageio[opencv]

This PR is also ready for review: https://github.com/nf-core/spatialvi/pull/123#issuecomment-3353657073 Adds the ability to process .tar files (frequently hosted by 10x instead of tar.gz) and fixes some meta handling that was preventing

resume

. Haven't properly tested the resume functionality, but seems like a good idea to change the handling of the meta paths in general. Thanks @Jonathan Manning for implementing the changes to the meta map handling! 👍

Hi all, I think we are now quite happy with the status of the dev branch, but one thing still missing is a full-size test data (see https://github.com/nf-core/spatialvi/issues/48). I was wondering whether anyone has experience working with 10X full-size datasets from other pipelines. Are there agreements with 10X for hosting one of their test datasets on object storage or something similar? Any input would be greatly appreciated!

Hi Spatialvi team, thank you for working on this pipeline! Is it possible to pin the Space Ranger version when running the spatialvi pipeline? For example, I want to use Space Ranger v2.1.0 instead of the current v3.1.3 to replicate a paper's result. Thanks!

Hi spatialvi team, The dev docs at https://nf-co.re/spatialvi/dev/ appear to be all blank, and the parameter module gives ``No nextflow_schema.json file found! It seems like there is no nextflow_schema.json file with parameters defined for this version of the pipeline. Try a newer version.` (See snapshots below) This persists across different browsers. The GitHub repo seems fine. Is it expected? Thanks!

Hello spatialvi team! I am trying to run the pipeline starting from spaceranger outputs However I am having trouble with the paths in the samplesheet.csv When my directory does not contain the proper files, I do get the fair error:

The specified spaceranger output directory for sample 'EX173983_R' does not contain all required files for either classic Visium: raw_feature_bc_matrix.h5, tissue_positions.csv, scalefactors_json.json, tissue_hires_image.png, tissue_lowres_image.png or Visium HD bin size 8: raw_feature_bc_matrix.h5, spatial/scalefactors_json.json, spatial/tissue_hires_image.png, spatial/tissue_lowres_image.png, spatial/tissue_positions.parquet.
ERROR ~ Pipeline failed. Please refer to troubleshooting docs: <https://nf-co.re/docs/usage/troubleshooting>

However if I include the necessary files, then the pipeline continues but soon stops with the error:

ERROR nextflow.processor.TaskProcessor - Error executing process > 'NFCORE_SPATIALVI:SPATIALVI:READ_DATA (1)'
Caused by:
  Not a valid path value: '250724-LW/EX173983_bis/'
Container:
  /beegfs/home/imaring/PROJECTS/PROMOTE_Spatial/work/singularity/community.wave.seqera.io-library-harmonypy_scanorama_gcc_gxx_pruned-95f731fde0b9ddef.img

I am not super expert in running nextflow pipelines and so I don't really know what to do to troubleshoot this. Hope you can help! Thanks and have a good day!

Hello spatialvi team. I am testing your pipeline for Visium HD data. I have a question: Is it possible to pass multiple hd bins resolutions per sample? At the moment I used this workaround of launching three different runs. Thank you

Hello spatial team, I had a small query. I saw your python script to calculate qc metrics, and i see that you have used a regex expression to get all hemoglobin genes. I was recently doing the same in my personal pipeline as i am learning to analyse spatial data and I saw that it doesnt specifically get only the hemoglobin related genes. I feel having a list of hemoglobin related genes explicitly mentioned would be a better options since they are very few of them themselves. This was my logic. Please correct me if am wrong here.

I have little experience with spatial datasets, but this tool looks promising. Once it is more mature, it may be an interesting addition to the pipeline? github.com/ScheWann/Loom