Skip to content
Tech News
← Back to articles

Reply to: Artefacts in single-cell mtDNA analyses misinform phylogenies

read original more articles
Why This Matters

This article highlights the potential for artefacts in single-cell mitochondrial DNA analyses to mislead phylogenetic interpretations, emphasizing the importance of accurate mutation calling and filtering strategies. These findings are crucial for improving the reliability of lineage tracing in cancer and other biological studies, ultimately benefiting both researchers and consumers of genomic data. Correcting these artefacts can lead to more precise insights into cellular evolution and disease progression.

Key Takeaways

(a) Example mtDNA mutations that show lineage specificity. The 1+-molecule mutations are represented in triangles with colors. The mutations with more molecules per cell are represented in black dots. (b-d) Related to Fig. 1g,h. (b) Reanalysis of the agreement of closeness (AOC) between ReDeeM and CRISPR lineage inference using ReDeeM pipeline by cell-ranger (bwa) as aligner (without mitochondrial genome masking) is computed with or without edge trimming. (c) AOC distribution compared to random reshuffled for panel-b. (d) AOC distribution using all mtDNA mutations called by 4 varieties of ReDeeM pipelines, where two aligners: bowtie2, and cell-ranger (bwa) and two consensus calling thresholds are tested. (e-f) Extended example from a different tumor sample is shown, related to Fig. 1g,h, The agreement of closeness (AOC) between ReDeeM and CRISPR lineage inference is computed across ReDeeM filter1 and filter2. 4 panels are in the same order. The AOC is computed for (1) mtDNA mutations using filter1, including 1+-molecule mutations, (2) mtDNA mutations using filter1, excluding 1+-molecule mutations (3) mtDNA mutations using filter2, including 1+-molecule mutations (4) mtDNA mutations using filter2, excluding 1+-molecule mutations. The regions with enhanced AOC when including 1-molecule mutations are highlighted. AOC distribution across different mtDNA filtering strategies compared to random reshuffled is shown in panel f.