M&M – an RNA-seq based pan-cancer classifier for pediatric tumors

Accurate diagnosis of rare tumors is a critical yet challenging task in pediatric oncology. Correctly identifying the type of tumor is essential for determining the most effective treatment plan and improving patient outcomes. To address this challenge, researchers at the Princess Máxima Center for Pediatric Oncology have developed M&M, a cutting-edge machine learning algorithm designed to enhance the diagnostic process for a wide range of tumor types, including rare ones.

What is M&M?

M&M stands for a pan-cancer ensemble-based machine learning algorithm. It utilizes RNA sequencing (RNA-seq) data to classify tumors with remarkable precision. RNA-seq is a powerful technique that captures the complete set of RNA transcripts present in a cell, providing a comprehensive snapshot of gene expression.

M&M framework

a) Schematic overview of the M&M framework, showing the separate Minority (left panel) & Majority classifier (right panel) machine learning workflows concerning feature selection, feature reduction, their down-sampling procedure, and their respective choice of algorithm. Note: the steps within the Majority classifier are not depicted in order, as cohort sub-setting takes place before feature selection. Classifier integration takes place after running the separate classifiers. The final probabilities were calculated by taking the average probability from the individual classifiers. If only one of the classifiers made a certain call, the final probability was divided by ten instead of averaged to penalize the classification label. b,c) Accuracy of separate Minority (red), Majority (blue), and integrated M&M classifiers (purple) for the tumor types (b) and subtypes (c) for different sample frequencies, determined in a ten-fold stratified cross-validation within the reference cohort.

How Does M&M Work?

M&M can classify:

  • 52 different tumor types with an impressive precision of around 99% and recall of about 80%.
  • 96 tumor subtypes with a precision of approximately 96% and recall of around 70%.

Precision refers to the algorithm’s ability to correctly identify positive cases out of all the cases it labels as positive, while recall measures the algorithm’s ability to identify positive cases out of all actual positive cases.

Handling Low-Confidence Classifications

For cases where the algorithm is less confident, M&M maintains high precision by considering the three highest-scoring labels. This approach ensures that even when the primary classification is uncertain, the algorithm still provides reliable options for further consideration.

Benefits of M&M’s Pan-Cancer Setup

One of the standout features of M&M is its pan-cancer setup. This means it uses a single classifier for all diagnostic samples, regardless of the tumor’s stage or the patient’s treatment status. This uniformity simplifies clinical implementation, as healthcare providers can use one consistent tool across various cases.

Comparing M&M to Existing Classifiers

M&M’s performance is comparable to existing tumor- and tissue-specific classifiers. However, its broad applicability and high precision for rare tumor types make it a valuable addition to the diagnostic toolkit. By increasing diagnostic accuracy, M&M has the potential to significantly improve treatment outcomes and quality of life for pediatric cancer patients.

Implications for Pediatric Oncology

The introduction of M&M in pediatric oncology diagnostics is a significant advancement. It promises to:

  • Enhance diagnostic accuracy for a wide range of tumor types, including rare and difficult-to-diagnose cancers.
  • Streamline the diagnostic process with a single, robust classifier.
  • Provide reliable results that help guide optimal treatment strategies.

M&M represents a leap forward in the fight against pediatric cancer. By leveraging advanced machine learning techniques and RNA-seq data, this algorithm offers a powerful tool for diagnosing rare tumors with high precision. Its implementation in clinical settings could lead to better diagnostic accuracy, more personalized treatment plans, and ultimately, improved survival rates and quality of life for young cancer patients. As technology continues to evolve, tools like M&M will play a crucial role in advancing pediatric oncology and providing hope for children and their families facing cancer diagnoses.

Wallis FSA, Baker-Hernandez JL, van Tuil M et al. (2024) M&M: An RNA-seq based Pan-Cancer Classifier for Pediatric Tumors. medRXiv [Epub ahead of print]. [article]

Leave a Reply

Your email address will not be published. Required fields are marked *

*

Time limit is exhausted. Please reload CAPTCHA.