Salas · Forensic science international. Genetics 2020 · Secondary computational re-analysis and methodological critique · n=?

Extraordinary claims require extraordinary evidence in asserted mtDNA biparental inheritance.

Level 5 - mechanism / opinion, no new human data

Secondary computational re-analysis and methodological critique of previously published datasets without new primary human data (Level 5 by design analogy).

PubMed 32330850 · doi:10.1016/j.fsigen.2020.102274 · record verified 2026-08-26

What was done

The authors re-analyzed original FASTQ sequencing files and published results from Luo et al. (PNAS), a study claiming evidence for biparental mitochondrial DNA (mtDNA) inheritance in humans. They evaluated potential analytical confounders—specifically nuclear mitochondrial DNA segments (NUMTs) and DNA rearrangements—assessed variant concordance across sequencing replicates from the same donors, and applied probabilistic and theoretical modeling to evaluate the proposed nuclear factor mechanisms.

What was found

The authors calculated that the probability of the biparental transmission events described by Luo et al. is extremely small, likely below 10^-37. Sequencing replicates from the same individuals displayed variant aberrations and discordances, pointing to methodological artifacts or extraneous DNA sources rather than genuine biology. Furthermore, theoretical modeling showed that the proposed nuclear factor would require an implausible level of complexity to prevent the linear accumulation of mixed mtDNA lineages across generations in the general population.

Why it matters

This critique reinforces the standard paradigm of strictly maternal mtDNA inheritance in humans by demonstrating that reported instances of biparental transmission likely reflect technical artifacts such as NUMTs. It underscores the necessity of stringent quality control and replication before accepting extraordinary genomic claims.

Limits

The study is a secondary computational re-analysis and commentary; no new primary patient samples or experimental laboratory validations were conducted. The abstract does not report the exact number of datasets, families, or sequencing runs re-analyzed.

Cited by