SMaHT Network Maps Somatic Mutations Across 19 Human Tissues in 150 Donors

Every human cell accumulates DNA changes over a lifetime — errors introduced during replication, repair failures, and mutagen exposures. These somatic mutations create a mosaic of genetically distinct cell populations within a single individual. While somatic mutations have been linked to cancer and other diseases, a systematic, tissue-wide reference for healthy individuals has been lacking, largely because such variants exist at very low allele frequencies and are technically challenging to detect.

The SMaHT Network addresses this gap through an ambitious, NIH-funded consortium involving dozens of institutions. The core strategy involves collecting tissue samples from 150 non-diseased human donors across 19 distinct tissue sites — spanning brain, heart, liver, blood, skin, colon, and more. This breadth allows researchers to compare mutational landscapes not only within individuals across tissues, but also across donors varying in age, sex, and ancestry.

A central innovation of SMaHT is the integration of multiple sequencing modalities. Bulk whole-genome sequencing at high depth, single-cell sequencing, and long-read technologies are deployed together to maximize sensitivity for low-frequency somatic variants including single nucleotide variants, insertions/deletions, copy number alterations, structural variants, and mobile element insertions. Paired computational pipelines are being developed and benchmarked specifically for somatic variant calling in non-cancer tissues.

Beyond cataloging mutations, SMaHT aims to characterize clonal expansions — instances where a single mutant cell has proliferated to form a measurable clone. Such expansions are well-documented in blood (clonal hematopoiesis) but are increasingly recognized in solid tissues including esophagus, liver, and brain. Understanding clonal dynamics across tissues could illuminate early steps toward malignancy and reveal tissue-specific aging mechanisms.

The resulting reference catalog will serve as a critical baseline for disease research: by knowing what somatic mutation burden and clonal structure look like in healthy tissue, investigators can better identify aberrant patterns in cancer, neurodegenerative disease, and cardiovascular conditions. Caveats include the cross-sectional nature of the cohort, which limits longitudinal inference, and the challenge of achieving sufficient sensitivity for very rare variants even with deep sequencing approaches.