The effects of bioinformatics preprocessing on cell-free DNA fragment analysis

Area of Science:

• Genomics
• Bioinformatics
• Cancer Biomarkers

Background:

• Cell-free DNA (cfDNA) shows promise as a biomarker for cancer diagnosis and monitoring.
• Standardization of cfDNA collection, extraction, and bioinformatics analysis pipelines is lacking.
• This study focuses on the impact of bioinformatics preprocessing on cfDNA analysis.

Purpose of the Study:

• To evaluate how different bioinformatics preprocessing steps affect genetic and epigenetic cfDNA features.
• To determine how these feature variations influence the ability to distinguish between healthy and cancer cfDNA samples.
• To identify optimal preprocessing strategies for enhanced cancer detection using cfDNA.

Main Methods:

• Analysis of low-pass whole-genome cfDNA sequencing data from 20 lung cancer and 20 healthy samples.
• Assessment of preprocessing variations: read trimming, alignment filtering, genome build selection, downsampling, and fragment size selection.
• Evaluation of cfDNA features: fragment size, end motifs, copy number alterations, and nucleosome footprints.

Main Results:

• Most cfDNA features are robust to common preprocessing choices but sensitive to sequencing coverage.
• Fragment length and end motifs are least affected by low coverage; nucleosome footprints are highly sensitive.
• In silico selection of shorter fragments enhances cancer signals but reduces overall data, with fragment end motif analysis showing the most benefit.
• Filtering alignments and genome build selection slightly improve cancer classification based on nucleosome coverage and copy number alterations.

Conclusions:

• Bioinformatics preprocessing settings have a minimal impact on cfDNA analysis.
• Synergistic effects between analytical approaches can be leveraged for improved cancer detection.
• Optimizing preprocessing and analytical strategies can enhance the utility of cfDNA as a cancer biomarker.