Abstract A016: ews-nf: A custom workflow for tumor cell annotation and analysis of single-cell RNA-sequencing of paired patient Ewing sarcoma specimens from the Sean Karl cohort
Bibliographic record
Abstract
Abstract Ewing sarcoma (EwS) is a fusion oncoprotein-driven primary bone cancer that demonstrates vast intra- and inter-tumoral heterogeneity. Tumor cell subpopulations, tumor progression, and therapeutic vulnerabilities of cell subpopulations are poorly understood. Paired patient samples (primary and metastatic disease or relapse) are rare at any one institution, and collaborative efforts are needed to address these pressing biologic questions. A national collaborative effort (the Sean Karl cohort) has been established to conduct single-cell RNAseq analyses of retrospective paired tumor samples from patients with EwS and serial samples due to metastasis or relapse using the GEM-X Flex Gene Expression protocol from 10x Genomics. Four analytic teams from multiple institutions will be performing custom downstream analyses to understand the therapeutic vulnerabilities of EwS cell subsets and discern immunobiologic dysfunction. To ensure reproducibility, all data pre-processing and common analyses, such as cell type annotation, are centralized using reproducible workflows developed by the Childhood Cancer Data Lab, a program of Alex’s Lemonade Stand Foundation. Gene expression is quantified using an open-source workflow, scpca-nf. The output from scpca-nf, which includes raw and normalized gene expression, dimensionality reduction, and annotation of non-malignant cells, is used as input to a custom Nextflow workflow, ews-nf, to annotate and analyze tumor cells in Ewing sarcoma samples. Tumor cells are annotated using two complementary methods: AUCell is used to evaluate expression of EwS-specific gene sets, and inferCNV is used to obtain a CNV profile for each cell by comparing potentially malignant cells to definitively non-malignant cells (e.g., immune cell types). Cells with high expression of EwS-specific gene sets and high CNV profiles, relative to immune cell types, are annotated as tumor cells. Tumor cells are further divided into EWS::FLI1 “low” and “high” cells based on expression of custom gene sets. All tumor cells are then analyzed using non-negative matrix factorization to identify and label recurrent gene expression programs found across all samples in the cohort. Processing and sequencing of samples are ongoing at the time of abstract submission. Ultimately, the output from ews-nf will be used to create a harmonized dataset to be shared with all four analytic teams. This harmonized dataset will contain the processed gene expression data, labeling of tumor cells and tumor cell states, and identification of recurrent gene expression programs. This enables all analytical teams to conduct downstream analysis using the same set of tumor cell annotations, making it easy for teams to compare results and draw conclusions. After completion of the study, the ews-nf workflow will be made publicly available to the research community. The processed gene expression data from the Sean Karl cohort, including the tumor cell annotations, will also be made available on the Single-cell Pediatric Cancer Atlas Portal for others to use in their own research. Citation Format: Allegra G Hawkins, Stephanie J Spielman, Joshua A Shapiro, Abbe Pannucci, Elina Mukherjee, Jessica Daley, Shireen Ganapathi, Elissa Boguslawski, Lea F Surrey, Patrick Azar, Filemon Dela Cruz, Jovana Pavisic, Emily Stockfisch, Azfar Neyaz, Ivy John, Jennifer Picarsic, Yutaro Tanaka, Riaz Gillani, Katherine A Janeway, Jaclyn N Taroni, Jessica Davis, Damon Reed, Adam Shlien, Theodore Laetsch, Rajen Mody, Elizabeth R Lawlor, Patrick Grohar, Anthony R Cillo, Kelly M Bailey. ews-nf: A custom workflow for tumor cell annotation and analysis of single-cell RNA-sequencing of paired patient Ewing sarcoma specimens from the Sean Karl cohort [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Discovery and Innovation in Pediatric Cancer— From Biology to Breakthrough Therapies; 2025 Sep 25-28; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2025;85(18_Suppl_2):Abstract nr A016.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.005 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.003 | 0.001 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.048 | 0.032 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".