Detecting and quantifying clonal selection in somatic stem cells

Area of Science:

• Somatic cell genetics
• Stem cell biology
• Computational biology

Background:

• Accumulation of DNA variants in somatic stem cells can alter tissue function.
• The mechanisms and timing of selection in homeostatic tissues remain poorly understood.
• Detecting selection in non-malignant tissues is crucial for understanding aging and disease initiation.

Purpose of the Study:

• To introduce SCIFER, a scalable computational method for identifying genetic selection in individual tissues.
• To infer stem cell dynamics, clone size, and age without prior knowledge of driver events.
• To analyze selection patterns in human bone marrow and brain tissues.

Main Methods:

• Development of SCIFER, a scalable computational method.
• Application of SCIFER to bulk whole-genome sequencing data from nonmalignant human bone marrow and brain.
• Analysis of DNA variant accumulation, selection, and clone dynamics in stem cell populations.

Main Results:

• Pervasive genetic selection was detected in both human bone marrow and brain tissues.
• Selected clones in hematopoiesis initiated uniformly throughout life, regardless of known drivers.
• Selected clones in the brain, including pre-malignant ones, originated predominantly in early life (childhood to young adulthood).

Conclusions:

• SCIFER is a broadly applicable tool for detecting and quantifying selection in renewing somatic tissues.
• Selection in human tissues occurs dynamically and varies between tissue types (e.g., bone marrow vs. brain).
• Understanding selection dynamics is key to comprehending tissue homeostasis, aging, and the early stages of diseases like glioma.