Published 2024 | Version v1.0
Dataset Open

Flow cytometry data from: "Computational measurable residual disease assessment in acute myeloid leukemia using mixture models"

  • 1. Department of Hematology, Amsterdam UMC, Vrije Universiteit Amsterdam, Amsterdam, The Netherlands
  • 2. Cancer Center Amsterdam, Imaging and Biomarkers, Amsterdam, The Netherlands
  • 3. Department of Hematology, Amsterdam UMC, Universiteit van Amsterdam, Amsterdam, The Netherlands
  • 4. Department of Epidemiology and Biostatistics, Amsterdam UMC, Vrije Universiteit Amsterdam, Amsterdam, The Netherlands
  • 5. Department of Hematology, Erasmus MC Cancer Institute, Rotterdam, The Netherlands
  • 6. HOVON Foundation, Rotterdam, The Netherlands

Description

This repository contains two large annotated flow cytometry datasets (BLAST110, LAIP29) for AML and healthy control patients used in Mocking et al.

The BLAST110 cohort was gated to identify immature myeloid blasts.

The LAIP29 cohort comprises different patients and was gated to also identify leukemic (LAIP+) populations.

Scripts for reproducing our analyses are available at: https://github.com/AUMC-HEMA/cMRD-manuscript

 

Datasets

Each dataset is structured contains the following two folders

  • /FCS/ contains raw .fcs files

  • /labels/ contains .csv files containing cell-specific annotations

The .csv files and .fcs files share a common parameter ("event_ID") which can be used to link the cell labels to each FCS file.

The LAIP29 contains an additional file folders /annotations/. This shows the characterization and abundance of each leukemic populations.

 

BLAST110

/BLAST110 contains FCS files from 110 different AML and healthy control samples:

  • 20 normal bone marrow (NBM) samples

  • 30 AML diagnosis ("Dx") samples

  • 90 AML follow up ("FU") samples

    • Collected at different timepoints after induction chemotherapy

    • Different levels of measurable residual disease (MRD)

The breakdown of this available in sample_info.csv

 

LAIP29

/LAIP29 contains FCS files from 29 different AML samples measured at diagnosis and follow up.

Each sample can have matching diagnosis and follow up timepoints. This information is available in sample_info.csv

 

Manual gating

BLAST110

In the BLAST110 cohort, was manually gated according to the following gating hierarchy:

  • WBC based on FSC-A, SSC-A

  • WBC/Singlets based on FSC-A, FSC-H

  • WBC/Singlets/CD45pos based on SSC-A, CD45

  • WBC/Singlets/CD45pos/Lymphocytes based on SSC-A, CD45

  • WBC/Singlets/CD45pos/CD45dim based on SSC-A, CD45

  • WBC/Singlets/CD45pos/CD45dim/CD34pos based on SSC-A, CD34

  • WBC/Singlets/CD45pos/CD45dim/CD117pos based on SSC-A, CD117

The "Blast" label combines the ouptut of CD34pos and CD117pos

 

LAIP29

The LAIP29 cohort was gated according conventional MRD analysis by experts and was annotated similarly to the BLAST110 cohort, with added detail for leukemic cells.

The LAIP column indicates whether a cell was gated as LAIP+.

Each unique leukemic population was given a unique identifier LAIP_ID. Because overlap between LAIPs is not taken into account during gating, each leukemic cell can belong to multiple LAIPs, resulting in a string (e.g., "1" for only LAIP 1, "1_2_3" for a cell positive for 3 LAIPs).

Files

BLAST110.zip

Files (17.0 GB)

Name Size Download all
md5:c1f33d32523843041ae9edba249fb24b
14.9 GB Preview Download
md5:46564b753a589ad0db264d6003de0522
2.1 GB Preview Download
md5:8bb147650ca3ab2d4091120369c52b7f
2.6 kB Preview Download

Additional details

Software

Repository URL
https://github.com/AUMC-HEMA/cMRD-manuscript
Programming language
Python, R