Analysis of scRNA-Sequencing Data of Early T cell Precursor Acute Lymphoblastic Leukemia

[eveegardner]

Proposed analysis

We propose to perform comprehensive cell type annotation and B-cell and T-cell receptor repertoire analysis on the single-cell RNA sequencing data from the Single Cell Profiling of Early T cell Precursor Acute Lymphoblastic Leukemia (Group ID: SCPCAB0007, Project ID: SCPCP000003) project. The goal of our analysis is to characterize cellular heterogeneity and immune signatures associated with treatment response and disease progression.

Scientific goals

1. Identify and characterize cellular populations across 59 patient samples and enhance cell type resolution by integrating surface protein expression data from CITE-seq data from 29 patient samples.
2. Define cell hierarchy and differentiation states to distinguish malignant cell populations from normal hematopoietic cells and identify potential cells of origin across leukemia subtypes.
3. Perform BCR/TCR annotation to analyze immune receptor repertoires to characterize T-cell and B-cell populations and compare immune signatures between treatment-responsive, treatment-refractory, and relapsed ETP and non-ETP ALL cases to identify prognostic immune biomarkers.

Methods or approach

Single-Cell RNA-Seq Processing and Cell Type Annotation

1. Data Processing Pipeline: Raw single-cell RNA-seq and CITE-seq data will be processed using the Alex's Lemonade Stand Foundation (ALSF) standardized scRNA-seq computational pipeline, which includes:
   • QC and filtering of low-quality cells and genes
   • Normalization and batch correction.
   • Dimensionality reduction and clustering analysis
   • Integration of RNA and surface protein expression data from CITE-seq.
2. Cell Type Classification:
   • Apply reference-based annotation using the HematopoieticImmuneCellAtlas.
   • Perform unsupervised clustering to identify novel cell states
   • Validate cell type assignments using canonical lineage markers and CITE-seq surface protein expression
   • Generate comprehensive cell type annotations across normal and malignant populations

BCR/TCR Repertoire Analysis

1. Immune Receptor Reconstruction: TRUST4 (TCR/BCR Tracking of Unpaired Single-cell Transcriptomes) will be employed to:
   • Reconstruct full-length TCR α/β and γ/δ chains from single-cell RNA-seq data
   • Perform V(D)J gene usage analysis and CDR3 sequence determination
2. Clonality and Diversity Analysis:
   • Calculate clonal expansion indices and diversity metrics
   • Identify shared TCR/BCR sequences across samples and conditions
   • Analyze clonal dynamics between diagnosis, treatment response, and relapse timepoints

Existing modules

This analysis will use the 30 cases of early T-cell precursor acute lymphoblastic leukemia (ETP ALL) stratified by treatment response and relapse status, 8 cases of acute myelogenous leukemia (AML), 11 cases of non-ETP T-ALL, 10 cases of T-myeloid mixed phenotype acute leukemia (MPAL).

Input data

SCPCAB0007 (Project ID: SCPCP000003), T-myeloid mixed phenotype acute leukemia (59 samples).

Scientific literature

1. Alex's Lemonade Stand Foundation Childhood Cancer Data Lab. (2023). Single-cell RNA-seq training modules. Training Modules for Computational Topics in Childhood Cancer Research. GitHub. https://github.com/AlexsLemonade/training-modules/tree/master/scRNA-seq
2. Human Cell Atlas Consortium. (2023). A single cell immune cell atlas of human hematopoietic system. Human Cell Atlas Data Portal. Retrieved from https://explore.data.humancellatlas.org/projects/cc95ff89-2e68-4a08-a234-480eca21ce79
3. Song, L., Cohen, D., Ouyang, Z. et al. TRUST4: immune repertoire reconstruction from bulk and single-cell RNA-seq data. Nat Methods 18, 627–630 (2021). https://doi.org/10.1038/s41592-021-01142-2

Other details

Computational Resources: We will utilize our local HPC to perform analysis and submit all documentation to GitHub, per the directions of ALSF.
Timeline: September - November 2025

[Jen-OMalley]

Hi @eveegardner. I hope you have been well! Thank you for interest in the OpenScPCA project. I wanted to share the contributor form. Filling out this form ensures you have agreed to the OpenScPCA terms and conditions and other policies. Once we receive this, we will get back to you here with any questions and/or next steps within three business days. We'll also start setting up an AWS account for you.

Please let me know if you have any questions. We look forward to working together!

[allyhawkins]

Hi @eveegardner, it's great to hear from you and thanks so much for your interest in OpenScPCA! I reviewed your analysis plan and have a few questions and comments to both give you some tips on working on your module, and to learn more about your planned approach.

General comments and questions about the proposal

• In your post, you noted that the data will be processed using the ALSF scRNA-seq pipeline. I just wanted to note that all of the data has already been processed through that pipeline and will be available to you as you work on this. The _processed.rds or _processed_rna.h5ad (and _processed_adt.h5ad) files will be the files you want to start with.
• For the reference-based annotation, what method do you plan to use?
• You also mention generating annotations across normal and malignant populations, but don't mention anything about how you will distinguish normal from malignant cells. For example, do you plan to use copy number inference, like inferCNV, or are there tumor-specific marker genes that you can use to pull out malignant cells?
• The analysis of the immune repertoire sounds very interesting, but I'm not sure it will really be feasible as part of OpenScPCA currently. It looks like TRUST4 is a tool to reconstruct the TCR/BCR repertoire from RNA-sequencing and requires the aligned BAM files as input. Unfortunately, we cannot share any raw sequencing data through this project. This particular data is available under access control through dbGaP, so if this is something you want to pursue as part of OpenScPCA, we could chat more about how that might work logistically while still adhering to our polices and terms of use. However, I would first focus on the cell type annotations for these samples.

Some notes on clustering

• In regards to using clustering to identify novel cell states, how do you plan to actually identify and label these cell states?
• Results from analysis like this can be highly dependent on clustering parameters, so it will be important to spend time optimizing your clustering parameters before attempting to "label" any of the clusters. I wanted to let you know that we have an R package called rOpenScPCA for your convenience to support common analyses in the project.
  • This package includes functionality for calculating clusters and evaluating their quality. Complete usage examples for rOpenScPCA clustering functions are available in the hello-clusters analysis module.
  • In general, if you want to use clusters as the basis for a set of annotations, it's going to be important to validate those clusters, so we hope this package can help you with that process!

Thanks for your interest in OpenScPCA, and let me know if I can clarify anything. I'm looking forward to chatting more about your proposal and helping you get started with contributions!

[eveegardner]

Hi @allyhawkins -- thank you for the clarification on these details, I very much appreciate this. We will work on a response to your questions and respond this week!

[Jen-OMalley]

Hi @eveegardner. Just sharing a quick reminder to submit the the contributor form when you have a chance. After we receive this, we can create your AWS account. Thank you again!

[eveegardner]

Thanks for the reminder, @Jen-OMalley ! We will fill this out today!

[eveegardner]

@Jen-OMalley submitted.

[Jen-OMalley]

Thanks @eveegardner. It looks like the form was submitted with the name and contact information for the PI in your lab? We will need a form from anyone who is directly working on the project to ensure you've personally agreed to the terms and policies. Would you be able to submit a form with your own information as well? Sorry about that! Once we receive your form, we can also create the AWS account using your information, if that's your preference.

Thank you so much!