Back

Analysis of Naturally Arising Simple Deletions in Single Human Colon Crypts with Age

Loh, Y. H. E.; Lieber, M. R.; Okitsu, T.; Okitsu, C.; Wlodarczyk, J.; Manojlovic, Z.; Hsieh, C.-L.

2026-10-06 genomics
10.64898/2026.09.30.755170 bioRxiv
Show abstract

Changes of DNA sequence in the human genome may contribute to human aging, but it has been difficult for the field to define the details of the contribution. Cancer cells have altered ratios of DNA damage and repair pathways, and thus it is inappropriate to utilize them as models for normal cells. Also, somatic mutations from each lineage cannot be clearly identified in cell populations with multiple lineages. Normal human colon crypts are one of the few mini-organ systems where monoclonal cell populations can be found for somatic mutation analysis. In a parallel study, we described complex insertion/deletion (indel) events in normal human colon. Here we focus on simple deletions that lack any nucleotide addition or adjacent base substitutions. We find that ~58% of the simple deletions are consistent with a strand slippage event, and the remainder provide insights into other important repair processes. Among the 4,264 >=2-nucleotide simple deletions, ~71% are most consistent with nonhomologous DNA end joining (NHEJ), with <1.5% of the events being more consistent with alternative end joining (aEJ). The incidence of NHEJ-type simple deletions is ~8-fold more frequent than the complex indels that also show NHEJ features. Similar to the complex indels, the simple deletions show a linear trend of event accumulation (positive slope) with an overall ~10-fold increase in absolute number across the human lifespan, 0.8 months to 90 years old, for individuals with no radiation or chemotherapy treatment. Interestingly, radiation treatment increases non-slippage deletions with little effect on slip deletions.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.