Prioritization of Fluorescence In Situ Hybridization (FISH) Probes for Differentiating Primary Sites of Neuroendocrine Tumors with Machine Learning

Lucas Pietan; Hayley Vaughn; James R Howe; Andrew M Bellizzi; Brian J Smith; Benjamin Darbro; Terry Braun; Thomas Casavant

doi:10.3390/ijms242417401

Back

Prioritization of Fluorescence In Situ Hybridization (FISH) Probes for Differentiating Primary Sites of Neuroendocrine Tumors with Machine Learning

Journal article

Open access

Peer reviewed

Prioritization of Fluorescence In Situ Hybridization (FISH) Probes for Differentiating Primary Sites of Neuroendocrine Tumors with Machine Learning

Lucas Pietan, Hayley Vaughn, James R Howe, Andrew M Bellizzi, Brian J Smith, Benjamin Darbro, Terry Braun and Thomas Casavant

International journal of molecular sciences, Vol.24(24), 17401

12/15/2023

DOI: 10.3390/ijms242417401

PMCID: PMC10743810

PMID: 38139230

Files and links (1)

url

https://doi.org/10.3390/ijms242417401View

Published (Version of record) Open Access

Abstract

Determining neuroendocrine tumor (NET) primary sites is pivotal for patient care as pancreatic NETs (pNETs) and small bowel NETs (sbNETs) have distinct treatment approaches. The diagnostic power and prioritization of fluorescence in situ hybridization (FISH) assay biomarkers for establishing primary sites has not been thoroughly investigated using machine learning (ML) techniques. We trained ML models on FISH assay metrics from 85 sbNET and 59 pNET samples for primary site prediction. Exploring multiple methods for imputing missing data, the impute-by-median dataset coupled with a support vector machine model achieved the highest classification accuracy of 93.1% on a held-out test set, with the top importance variables originating from the FISH probe. Due to the greater interpretability of decision tree (DT) models, we fit DT models to ten dataset splits, achieving optimal performance with k-nearest neighbor (KNN) imputed data and a transformation to single categorical biomarker probe variables, with a mean accuracy of 81.4%, on held-out test sets. and variables ranked as top-performing features in 9 of 10 DT models and the full dataset model. These findings offer probabilistic guidance for FISH testing, emphasizing the prioritization of the , , and FISH probes in diagnosing NET primary sites.

Machine Learning

fluorescence in situ hybridization

neuroendocrine tumor

model

imputation

biomarker

Details

Title: Subtitle: Prioritization of Fluorescence In Situ Hybridization (FISH) Probes for Differentiating Primary Sites of Neuroendocrine Tumors with Machine Learning
Creators: Lucas Pietan - University of Iowa
Hayley Vaughn - University of Iowa
James R Howe - University of Iowa
Andrew M Bellizzi - University of Iowa
Brian J Smith - University of Iowa
Benjamin Darbro - University of Iowa
Terry Braun - University of Iowa
Thomas Casavant - University of Iowa
Resource Type: Journal article
Publication Details: International journal of molecular sciences, Vol.24(24), 17401
DOI: 10.3390/ijms242417401
PMID: 38139230
PMCID: PMC10743810
NLM abbreviation: Int J Mol Sci
ISSN: 1661-6596
eISSN: 1422-0067
Grant note: T32 GM 008629 / NIH HHS P50 CA174521 / NCI NIH HHS
Language: English
Date published: 12/15/2023
Academic Unit: Roy J. Carver Department of Biomedical Engineering; Electrical and Computer Engineering; Stead Family Department of Pediatrics; Pathology; Biostatistics; Medical Genetics and Genomics; Surgery; Holden Comprehensive Cancer Center
Record Identifier: 9984532057102771

Metrics

19 Record Views

1 Times Cited - Web of Science