RNA sequencing and swarm intelligence-enhanced classification algorithm development for blood-based disease...

Area of Science:

• Biotechnology
• Bioinformatics
• Molecular Diagnostics

Background:

• Blood-based biomarkers offer potential for revolutionary disease diagnostics and minimally invasive therapy monitoring.
• Selecting relevant biomarkers from liquid biosources presents a significant challenge.
• The thromboSeq pipeline was previously developed for RNA sequencing and cancer classification using platelet RNA and swarm intelligence.

Purpose of the Study:

• To provide a comprehensive wet-lab and dry-lab protocol for generating platelet RNA sequencing libraries.
• To detail the development of swarm intelligence-enhanced machine learning algorithms for classification based on platelet RNA.
• To enable the scientific community to utilize platelet RNA for diagnostic algorithm development.

Main Methods:

• Wet-lab protocol: platelet RNA isolation, mRNA amplification, and library preparation for next-generation sequencing.
• Dry-lab protocol: automated FASTQ pre-processing, gene count quantification, quality control, data normalization, and correction.
• Development of swarm intelligence-enhanced support vector machine (SVM) algorithms for classification.

Main Results:

• The protocol allows for platelet RNA profiling starting from 500 pg of RNA.
• Enables automated and optimized selection of biomarker panels.
• Wet-lab protocol completion in 5 days; algorithm development in 2 days, contingent on computational resources.

Conclusions:

• This protocol facilitates platelet RNA analysis for diagnostic algorithm development.
• It empowers researchers with a reproducible method for biomarker discovery from blood.
• The approach supports the advancement of minimally invasive diagnostic tools.