Nano Random Forests to mine protein complexes and their relationships in quantitative proteomics data

Area of Science:

• Proteomics
• Systems Biology
• Bioinformatics

Background:

• Quantitative proteomics generates vast datasets valuable for understanding protein function.
• Multiprotein complexes exhibit consistent trends in proteomics experiments, offering biological insights.
• Identifying small complexes and their relationships is challenging due to data complexity and the need for large training sets.

Purpose of the Study:

• To evaluate the capability of the Random Forests (RF) machine learning algorithm in detecting small protein complexes and their relationships using quantitative proteomics data.
• To demonstrate that RF can identify distinguishable signatures for small protein complexes, challenging the assumption of needing large training datasets.
• To apply the developed approach to identify protein complexes and functional links within mitotic chromosome proteomics data.

Main Methods:

• Utilized the Random Forests (RF) machine learning algorithm.
• Developed a complex-oriented RF approach (NanoRF) for analyzing quantitative proteomics data.
• Tested the method with simulated and real proteomics data, including wild-type and knockout mitotic chromosome experiments.

Main Results:

• RF successfully detected small protein complexes and their interrelationships in quantitative proteomics data.
• Identified several protein complexes within mitotic chromosome data, with associated proteins suggesting novel functional links.
• Revealed known subunit interdependences within the kinetochore and a novel link between the inner kinetochore and condensin.
• Confirmed the independence of ribosomal proteins from kinetochore subcomplexes.

Conclusions:

• The complex-oriented RF (NanoRF) approach effectively integrates proteomics data to uncover subtle protein relationships.
• NanoRF demonstrates the ability to detect small protein complexes and their functional associations, even with limited training data.
• The study highlights potential novel functional links and interdependencies among proteins in mitotic chromosomes, advancing systems biology research.