Dataset Information

Rapid Epigenomic Classification of Acute Leukemia

ABSTRACT: Acute leukemias (AL) are aggressive blood cancers that require precise molecular classification and urgent treatment. However, standard-of-care diagnostic tests are time and resource intensive and do not capture the full spectrum of AL heterogeneity. Here, we developed a machine learning framework to classify AL using genome-wide DNA methylation profiling. We first assembled a large reference cohort (n=2,540 samples) and defined 38 distinct methylation classes across AL lineages and age groups. A subset of methylation classes mirrored established AL categories defined by genetics, such as AML with CBFB::MYH11 and B-ALL with ETV6::RUNX1, while others revealed epigenetic heterogeneity not captured by standard-of-care testing alone, including four methylation classes associated with NPM1-mutant AML (HOXA/B-activated) and four with KMT2A-rearranged AML (HOXA-activated). Using this reference, we next developed a deep neural network model (MARLIN) for methylation-based AL classification from extremely sparse data, and applied this for rapid analysis of clinical samples profiled by nanopore sequencing. In a retrospective AL cohort, MARLIN-based classifications were concordant with standard-of-care diagnoses in 25/26 (96.2%) cases with high-confidence prediction scores, including samples of diverse lineages and molecular subtypes. In five additional patients presenting with suspected AL, we performed nanopore sequencing and MARLIN classification of clinical blood and bone marrow samples in real-time, typically achieving an accurate methylation class prediction in less than two hours from sample receipt and less than one hour of sequencing time. In summary, we present a DNA methylation-based machine learning framework for rapid AL classification in the clinic that can complement and enhance standard-of-care diagnostics.

ORGANISM(S): Homo sapiens

PROVIDER: GSE280090 | GEO | 2025/06/16

REPOSITORIES: GEO

ACCESS DATA

Dataset's files

Source:

			Action	DRS
		Other
	GSE280090_RAW.tar	Raw
	filelist.txt	Txt

Items per page:

1 - 3 of 3

Similar Datasets

Project description:BackgroundAcute myeloid leukemia (AML) is characterized by uncontrolled proliferation of malignant hematopoietic cells in the bone marrow that are arrested in differentiation. In AML pathogenesis, hematopoietic progenitor cells acquire multiple genetic aberrations often occurring in the same set of genes that ultimately lead to malignant transformation. In the WHO classification 2016, six AML classes with different prognosis are identified by chromosomal translocations measured by standard cytogenetics. All balanced translocations produce fusion genes, except for t(3;3)/inv(3;3), which lead to overexpression of MECOM/EVI1 associated with poor prognosis. For accurate risk assessment of cytogenetically normal AML, four genes need to be screened to distinguish AML with mutated NPM1 in the absence of FLT3 mutations and AML with bi-allelic CEBPA mutations, which have a favorable prognosis, and AML with RUNX1 mutations which have a poor prognosis. Recently, a full genomic classification system has been proposed by Papaemmanuil et al. (NEJM 2016). In this system, the same six classes with chromosomal translocations are identified by standard cytogenetics as well as five additional classes characterized by genetic mutations in 14 different genes. AimThe aim of the study was to investigate whether whole transcriptome RNA-sequencing (RNAseq) can be used as single technology for classification of AML. MethodA panel of 100 AML were analyzed and a bio-informatics pipeline called HAMLET (Human AML Expedited Transcriptomics) was developed in which 4 modules are integrated to detect (1) fusion genes, (2) small variants in 13 recurrently mutated genes, (3) internal tandem duplications in FLT3 (FLT3-ITD) and partial tandem duplications in KMT2A (KMT2A-PTD) and (4) overexpression of MECOM/EVI1. All mutations that were called by HAMLET were validated by diagnostic data as available for all 100 AML or targeted PCR followed by Sanger or next generation sequencing on all positive AML and at least an equal number of negative cases. Results & discussionThe data showed that HAMLET accurately identified all genetic aberrations with high sensitivity and specificity. In addition, in 7 AML, fusion transcripts were detected that are not measured by standard cytogenetics including three cases that lack any class-defining lesion according to Papaemmanuil et al. Moreover, overexpression of MECOM/EVI1 was detected in two AML with inv(3) as well as in five cases without inv(3)/t(3;3), and a gene signature was measured to distinguish AML with CEBPA mutations with favorable prognosis. In conclusion, HAMLET provides a comprehensive and flexible pipeline for RNAseq analysis to retrieve all relevant information for current classification of AML as well as additional information that may improve classification in the future.

Dataset Information

Rapid Epigenomic Classification of Acute Leukemia

Dataset's files

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets