dendroulab / panpipes-tutorials Goto Github PK
View Code? Open in Web Editor NEWtutorials to run panpipes
License: BSD 3-Clause "New" or "Revised" License
tutorials to run panpipes
License: BSD 3-Clause "New" or "Revised" License
Got the following issue during the visualisation tutorial Error in library(cowplot) : there is no package called 'cowplot' \
and in the installation guide there is no indication to install the package so maybe should be included there.
what to do when the pipeline breaks mid-run: sometimes the pipeline will stop because a parameter is wrong in the config file or because of any small issue in the yml such as the wrong path for a certain file. We recommend checking the log file for information on what exactly failed. After fixing the issue we recommend cleaning the project folder of any intermediate files, to allow a proper re-run.
The four ingestion tutorials have a "longer" version of the config file yml.
we need to align them to the new config file here https://github.com/DendrouLab/panpipes/blob/main/panpipes/panpipes/pipeline_ingest/pipeline.yml
files to modify:
Hello,
I run the command: panpipes ingest config and dot this error:
RuntimeError: This version of jaxlib was built using AVX instructions, which your CPU and/or operating system do not support. You may be able work around this issue by building jaxlib from source.
Thank you,
Just as a reminder for myself to adapt the prefilled pipeline.yml
for the preprocess tutorial as soon as the respective PR is merged.
pipeline.yml
sections shown in the tutorial itselfScatter color function is not working and it can't recognize any inputs (even a hex colour code).
Tried changing the pair_scatters_markers.csv colour indication to rna:douplet_scores
, douplet_scores
and a hex code
and all of them errored
The first two features are read fine but not the third one.
Tested in python environment with mu.pl.scatter
function and same error appeared.
Also tested following error suggestion of doubling the rna:rna:doublet_scores
and same issue appeared again.
Additionally tested with the sc.pl.scatter
function in a python environment using simple gene markers and color as doublet_scores
and it worked so it must be a muon issue.
Additionally no instruction on how to accurately create the input file for scatter plots on the visualisation workflow or the gene format list.
Clustering issue for atac as it requires a signac_norm
layer however during the preprocessing steps atac is normalised via normalize: log1p
So as a result we have a logged_counts layer rather than a signac_norm one which causes the clustering pipeline tutorial to fail.
As discussed, we want to remove the specified conda environment paths from the prefilled tutorial pipeline.yml
files, as they cause the workflow to crash when run by the user. Specifically, we need to do the following:
Pipeline.yml from the tutorial has an option in the total VI filter_adt_outliers
however that does not allow it to run properly. The tutorial git yml (pipeline_integration.yml) shows an outdated version, however the panpipes git yml has an updated version which fixes the issue by replacing it with filter_prot_outliers
. So tutorial yml should be updated as well
Vis python script does not make the scatters directory that is required in the tutorial pipeline.yml so it errors.
So for visualisation to run user needs to manually create a new directory inside /vis/data called scatters.
Downloaded files from tutorial paired_scatter, custom_markers indicate prot_CD8, however the teaseq_clusterd.h5mu file created during clustering presents with prot_CD8a so during visualisation it can't find the correct prot_CD8.
Protein name in the downloaded files from the visualisation tutorials has to be changed to prot_CD8a
For the section "Filtering data with panpipes - running the pipeline"
The pre-filled pipeline.yml file (here) contains an error on the specificed ATAC normalisation method, and needs to be updated.
# ATAC preprocessing steps
# ------------------------
atac:
binarize: False
normalize: TFIDF #"log1p" or "TFIDF"
The figures generated using this pre-filled pipeline.yml do not match the PCA plot images posted in the documentation. This is because the pre-filled YAML file uses TFIDF
, but the plot images were generated using log1p
.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.