Learning predictive models for combinations of heterogeneous proteomic data sources

TL;DR

This paper explores combining heterogeneous proteomic data sources, demonstrating that specialized fusion models outperform individual models in classifying pancreatic cancer data.

Contribution

It introduces a class of model fusion methods tailored for heterogeneous proteomic datasets, improving classification performance.

Findings

Fusion models outperform individual models on combined datasets.

Models effective on single data sources often fail on combined data.

Proposed fusion methods leverage differences to enhance classification accuracy.

Abstract

Multiple technologies that measure expression levels of protein mixtures in the human body offer a potential for detection and understanding the disease. The recent increase of these technologies prompts researchers to evaluate the individual and combined utility of data generated by the technologies. In this work, we study two data sources to measure the expression of protein mixtures in the human body: whole-sample MS profiling and multiplexed protein arrays. We investigate the individual and combined utility of these technologies by learning and testing a variety of classification models on the data from a pancreatic cancer study. We show that for the combination of these two (heterogeneous) datasets, classification models that work well on one of them individually fail on the combination of the two datasets. We study and propose a class of model fusion methods that acknowledge the differences and try to reap most of the benefits from their combination.