Inference of trajectory presence by tree dimension and subset specificity by subtree cover

Area of Science:

• Computational Biology
• Systems Biology
• Genomics

Background:

• Cellular differentiation involves complex dynamic transitions between states.
• Single-cell biology enables trajectory inference to model these progressions.
• Current trajectory inference methods lack statistical rigor for noisy data.

Purpose of the Study:

• To introduce a statistically robust method for assessing trajectory presence in multivariate data.
• To develop a measure for quantifying tissue specificity of gene expression dynamics.
• To evaluate the conservation of pathway dynamics specificity in human and mouse development.

Main Methods:

• Introduced a tree dimension test using Euclidean minimum spanning tree measures and a null distribution.
• Developed a subset specificity measure based on minimum subtree cover.
• Applied methods to simulated and real single-cell RNA sequencing datasets.

Main Results:

• The tree dimension test effectively assesses trajectory significance, outperforming existing statistics.
• Tissue specificity of pathway gene expression dynamics is conserved between human and mouse development.
• Signal transduction pathways show high tissue specificity, while genetic information processing pathways show low specificity.

Conclusions:

• The tree dimension test provides a statistically sound approach to trajectory inference.
• The subset specificity measure offers insights into conserved developmental dynamics.
• These methods enable prioritization of cellular dynamics and pathways in biological systems.