NGS IEG: March 19th - Classifying B-ALL Genomic Subtypes
Greetings!
Dust off your fusion genes and aneuploidies — it’s time for our March NGS meeting. This month we’re hosting NGS IEG’s most helpful organizer, John Lin! John will be stepping into the world of long-read transcriptomics and B-ALL subtype classification. See the talk details below.
Title: A Classification Pipeline for B-ALL Genomic Subtypes using mRNA-sequencing
Abstract: The clinical classification of B-cell acute lymphoblastic leukemia (B-ALL) genomic subtypes (gross aneuploidies and fusion oncogenes) has significant prognostic implications. Previously, we showed a partial least squares regression-support vector machine (PLSR-SVM) composite model classified broad B-ALL genomic subtypes based on Oxford Nanopore Technologies whole-transcriptome sequencing (ONT-WTS). I present an enhanced B-ALL classification pipeline (leveraging a direct B-ALL fusion caller and aneuploid classifier), resulting in more granular B-ALL genomic subtyping. The PLSR-SVM model was trained on 1,036 transcriptomes, derived from Illumina short-read mRNA-sequencing, and 523 transcriptomes, derived from ONT-WTS. Gene fusion detection corrects genomic subtypes including ETV6-RUNX1, TCF3-PBX1 and MEF2Dr, while the aneuploid classifier correctly classifies 92% near haploid and high hyperdiploid subtypes. Overall, the updated B-ALL classification pipeline results in 87% accuracy, spanning subtypes: ETV6-RUNX1, Ph/Ph-like, KMT2Ar, TCF3-PBX1, near haploid, low hypodiploid, high hyperdiploid, DUX4r, MEF2Dr, and ZNF384r. Integrating direct fusion calls and specific aneuploid classifications results in improved subtype classification. Future work involves refinement of the model’s hyperparameters and validation on external datasets.
Presenter’s bio: John is a doctoral candidate in Jeremy Wang’s Lab at the University of North Carolina at Chapel Hill in the Bioinformatics and Computational Biology program. He has a B.S. and M.S. in Biomedical Engineering from Northwestern University and Columbia University, respectively. Previously, he worked for Epic, IBM Watson Health, and ConcertAI utilizing real-world, clinical data for population health management and secondary use for biomedical research. Currently, he studies computational methods using long-read sequencing to detect and classify genomic subtypes of pediatric leukemias with a focus on B-cell acute lymphoblastic leukemia.
Date:
March 19th, 2026
Schedule:
Lunch/social - 12:00 PM to 12:30 PM
Presentation - 12:30 PM to 1:30 PM
Social hour - 1:30 PM to 2:00 PM
Format: Meeting will be held both virtually and in-person. Lunch will be provided for in-person attendees.
Add to calendar:
https://calendar.app.google/YUgHBTzabZM3zuhDA
Event Address:
North Carolina Biotechnology Center
15 T.W. Alexander Drive
Research Triangle Park,
NC 27709
Zoom link: https://ncbiotech.zoom.us/j/86440658342?pwd=MU9IckpYQUhHTnV6NE1UcXUvYnNXUT09
Meeting ID: 864 4065 8342
Passcode: 583621
Other NGS IEG info/links:
● The NextGen Sequencing Intellectual Exchange Group (NGS IEG) provides a forum to share methods and technologies driving biotechnology research in North Carolina from both academic and private research groups. NGS IEG provides an excellent networking opportunity for biotech leaders, academics, and students interested in networking with local RTP biotech. We gather industry leaders from the RTP area to share, discuss and review their new technologies, research, and big-data platforms in a casual environment with lunch provided. Related NGS IEG links below: