Explore consensus cell types in brain and CNS tumors #1036

allyhawkins · 2025-02-11T22:22:32Z

Purpose/implementation Section

Please link to the GitHub issue that this pull request addresses.

What is the goal of this pull request?

Here I'm summarizing the findings for consensus cell types in all samples that have a brain/cns diagnosis. This includes samples from SCPCP000001, SCPCP000002, SCPCP000009, SCPCP000010, and SCPCP000021. I did not include the samples that belong to multiplexed libraries.

Briefly describe the general approach you took to achieve this goal.

The overall flow of the notebook is similar to the osteo notebook, which means a lot of the setup code is copied from that notebook. First we look at the top cell types across all samples and then facet that by project id and subdiagnosis. I also look at the total immune cell classification and then specifically T cell populations vs. macrophages/ monocyte lineage.
As with osteo, there's variation across all the samples in terms of how many cells are classified and generally this is not project specific. I think the most interesting thing is looking at the diagnosis groups, since there appears to be a difference in T cell infiltrate and macrophage infiltrate in LGG vs. HGG. Obviously this is all qualitative, but I do think this could be something to pursue for showing in the paper.
For looking at sub diagnosis, I looked at the most represented diagnoses first and then in the last plot I grouped everything so it's either LGG or HGG.

Note that in working on this I moved the functions I used for plotting in the osteo notebook to their own script to source in.

If known, do you anticipate filing additional pull requests to complete this analysis module?

Yes.

Results

What types of results does your code produce (e.g., table, figure)?

Here's a copy of the rendered report:
04-brain-and-cns-consensus-celltypes.nb.html.zip

What is your summary of the results?

As mentioned above I think the most interesting thing is the qualitative difference between T cell and macrophage infiltrate. It looks like Glioblastoma have a higher percentage of T cells and much less macrophages while the other types tend to mostly have a higher percentage of macrophages. This trend appears to be more generally divide HGG and LGG too, but again this is qualitative.

Author checklists

Check all those that apply.
Note that you may find it easier to check off these items after the pull request is actually filed.

Analysis module and review

This analysis module uses the analysis template and has the expected directory structure.
The analysis module README.md has been updated to reflect code changes in this pull request.
The analytical code is documented and contains comments.
Any results and/or plots this code produces have been added to your S3 bucket for review.

Reproducibility checklist

Code in this pull request has been added to the GitHub Action workflow that runs this module.
The dependencies required to run the code in this pull request have been added to the analysis module Dockerfile.
If applicable, the dependencies required to run the code in this pull request have been added to the analysis module conda environment.yml file.
If applicable, R package dependencies required to run the code in this pull request have been added to the analysis module renv.lock file.

jaclyn-taroni

The overall plotting strategy looks good, but I have some comments/questions about various categorizations.

jaclyn-taroni · 2025-02-12T15:32:35Z

analyses/cell-type-consensus/exploratory-notebooks/04-brain-and-cns-consensus-celltypes.Rmd

+Based on this let's make the following groups: 
+
+- Low-grade glioma: Any samples labeled with "Low-grade glioma" or are part of `SCPCP000002` which are categorized as LGG tumors. 
+- High-grade glioma: Any samples labeled with "High-grade glioma" or are part of `SCPCP000001` which are categorized as HGG tumors. 


This should include Diffuse midline glioma as well (another source).

jaclyn-taroni · 2025-02-12T15:37:49Z

analyses/cell-type-consensus/exploratory-notebooks/04-brain-and-cns-consensus-celltypes.Rmd

+  "macrophage", 
+  "monocyte",
+  "mononuclear phagocyte", 
+  "dendritic cell"


Could you only get a myeloid dendritic cell out of the consensus labels? https://pmc.ncbi.nlm.nih.gov/articles/PMC5032838/

jaclyn-taroni · 2025-02-12T15:43:50Z

analyses/cell-type-consensus/exploratory-notebooks/04-brain-and-cns-consensus-celltypes.Rmd

+  stringr::str_subset("T")
+
+# macro/mono lineage 
+macro_mono_cells <- c(


You could probably call this myeloid and include granulocytes with little to no impact on the results?

analyses/cell-type-consensus/exploratory-notebooks/utils/plotting-functions.R

jashapiro

This looks good to me, with @jaclyn-taroni's comments addressed. One little additional comment.

jashapiro · 2025-02-12T15:56:25Z

analyses/cell-type-consensus/exploratory-notebooks/04-brain-and-cns-consensus-celltypes.Rmd

+```{r}
+# t cell subtypes
+t_cells <- unique_celltypes |> 
+  stringr::str_subset("T")


This seems potentially brittle. It looks like maybe you can use "T cell"?

Co-authored-by: Jaclyn Taroni <19534205+jaclyn-taroni@users.noreply.github.com>

allyhawkins · 2025-02-12T17:09:16Z

Thanks @jaclyn-taroni and @jashapiro. Based on your comments, I added "Diffuse midline glioma" to the HGG group and then I adjusted the macro/mono group, which I'm now labeling as myeloid. In doing that I ended up removing "dendritic cell", since one of the possible dendritic cells is of lymphoid lineage (see plasmacytoid dendritic cell or pDCs). Note that I checked and only one of the cells across all the samples used here was classified as pDC instead of DC, but I still kept it off. I did add "neutrophils" and "myeloid leukocyte" to the myeloid group, but I didn't add granulocytes (#1036 (comment)) since I think you actually meant neutrophils? I looked through the list of identified cell types here, not just the top ones shown in the plot, and those were all the myeloid cell types.

Here's the updated report.
04-brain-and-cns-consensus-celltypes.nb.html.zip

jaclyn-taroni · 2025-02-12T17:13:34Z

I did add "neutrophils" and "myeloid leukocyte" to the myeloid group, but I didn't add granulocytes (#1036 (comment)) since I think you actually meant neutrophils?

I meant granulocytes in the general sense without any consideration for whether or not a cell could get annotated as an eosinophil, for example. Put another way, I wasn't paying attention to the potential consensus cell labels. If only neutrophil is possible, that works for me.

jaclyn-taroni

I'm going to approve, since I don't think I need to see this again, but the text in the section about myeloid vs. T lineage should get rewritten before this is merged.

analyses/cell-type-consensus/exploratory-notebooks/04-brain-and-cns-consensus-celltypes.Rmd

jaclyn-taroni · 2025-02-12T17:15:13Z

analyses/cell-type-consensus/exploratory-notebooks/04-brain-and-cns-consensus-celltypes.Rmd

+It looks like there may be some differences between T cell and macrophage infiltrate based on the tumor type. 
+Here we will look specifically at all T cells and anything in the macrophage/monocyte lineage. 
+I'm including any cell types that are in our list of possible cell types and a child of [mononuclear phagocyte](https://www.ebi.ac.uk/ols4/ontologies/cl/classes/http%253A%252F%252Fpurl.obolibrary.org%252Fobo%252FCL_0000113?lang=en) in the macro/mono group. 
+We should also note that "macrophage" here probably refers to microglia instead. 


This text could use a rewrite now.

analyses/cell-type-consensus/exploratory-notebooks/04-brain-and-cns-consensus-celltypes.Rmd

jaclyn-taroni · 2025-02-12T17:16:31Z

analyses/cell-type-consensus/exploratory-notebooks/04-brain-and-cns-consensus-celltypes.Rmd

+- As expected based on the previous plot, glioblastoma appears to have greater T cell infiltrate than macrophages across all samples. 
+- There appears to be a handful of samples in the "High-grade glioma" group that also have high T cell infiltrate. 
+Notably, glioblastoma is a type of HGG. 
+- Generally, the T cell population seems lowest in LGG. 
+Ganglioglioma also has fewer samples with high percentage of T cell infiltrate and this is also a LGG. 
+- It looks like about 50% of the pilocytic astrocytomas have more T cell infiltrate than macrophages. 


Same comment about lightly rewriting this section.

Co-authored-by: Jaclyn Taroni <19534205+jaclyn-taroni@users.noreply.github.com>

allyhawkins · 2025-02-12T17:34:17Z

Just noting that I updated the text in b73b573.

allyhawkins added 3 commits February 11, 2025 16:12

move plotting functions to utils

bc3d246

notebook looking at brain/ cns samples

09e339c

update readme

edbc80b

allyhawkins requested a review from jaclyn-taroni as a code owner February 11, 2025 22:22

allyhawkins requested a review from jashapiro February 11, 2025 22:23

jaclyn-taroni reviewed Feb 12, 2025

View reviewed changes

jashapiro approved these changes Feb 12, 2025

View reviewed changes

allyhawkins and others added 2 commits February 12, 2025 10:47

Apply suggestions from code review

e8e542c

Co-authored-by: Jaclyn Taroni <19534205+jaclyn-taroni@users.noreply.github.com>

adjust HGG and myeloid groupings

e7a1601

allyhawkins requested a review from jaclyn-taroni February 12, 2025 17:09

jaclyn-taroni approved these changes Feb 12, 2025

View reviewed changes

allyhawkins and others added 2 commits February 12, 2025 11:28

Apply suggestions from code review

b9d649c

Co-authored-by: Jaclyn Taroni <19534205+jaclyn-taroni@users.noreply.github.com>

some rewording

b73b573

allyhawkins merged commit 8bbaf7a into AlexsLemonade:main Feb 12, 2025
3 checks passed

allyhawkins deleted the allyhawkins/brain-cns-consensus branch February 12, 2025 17:45

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Explore consensus cell types in brain and CNS tumors #1036

Explore consensus cell types in brain and CNS tumors #1036

allyhawkins commented Feb 11, 2025 •

edited

Loading

jaclyn-taroni left a comment

jaclyn-taroni Feb 12, 2025

jaclyn-taroni Feb 12, 2025

jaclyn-taroni Feb 12, 2025

jashapiro left a comment

jashapiro Feb 12, 2025

allyhawkins commented Feb 12, 2025

jaclyn-taroni commented Feb 12, 2025

jaclyn-taroni left a comment

jaclyn-taroni Feb 12, 2025

jaclyn-taroni Feb 12, 2025

allyhawkins commented Feb 12, 2025

Explore consensus cell types in brain and CNS tumors #1036

Explore consensus cell types in brain and CNS tumors #1036

Conversation

allyhawkins commented Feb 11, 2025 • edited Loading

Purpose/implementation Section

Please link to the GitHub issue that this pull request addresses.

What is the goal of this pull request?

Briefly describe the general approach you took to achieve this goal.

If known, do you anticipate filing additional pull requests to complete this analysis module?

Results

What types of results does your code produce (e.g., table, figure)?

What is your summary of the results?

Author checklists

Analysis module and review

Reproducibility checklist

jaclyn-taroni left a comment

Choose a reason for hiding this comment

jaclyn-taroni Feb 12, 2025

Choose a reason for hiding this comment

jaclyn-taroni Feb 12, 2025

Choose a reason for hiding this comment

jaclyn-taroni Feb 12, 2025

Choose a reason for hiding this comment

jashapiro left a comment

Choose a reason for hiding this comment

jashapiro Feb 12, 2025

Choose a reason for hiding this comment

allyhawkins commented Feb 12, 2025

jaclyn-taroni commented Feb 12, 2025

jaclyn-taroni left a comment

Choose a reason for hiding this comment

jaclyn-taroni Feb 12, 2025

Choose a reason for hiding this comment

jaclyn-taroni Feb 12, 2025

Choose a reason for hiding this comment

allyhawkins commented Feb 12, 2025

allyhawkins commented Feb 11, 2025 •

edited

Loading