Revealing the platelet transcriptome using Siphon-seq: a sensitive, purity-aware method for RNA-sequencing
In the authors' words
Platelet transcriptomics is limited by the exceptionally low RNA content of platelets and by disproportionate contributions from contaminating cells to bulk RNA-sequencing libraries. We developed Siphon-seq, an approach that deliberately overloads independently barcoded single-cell partitions to increase aggregate platelet RNA input, then computationally identifies and removes partitions containing contaminating RNA after sequencing. Applied to 34 healthy human donors, Siphon-seq increased platelet-marker representation, depleted leukocyte and erythrocyte transcripts, and removed contamination-associated inter-donor transcriptional structure. We defined a 3,436-gene core platelet transcriptome that was strongly related to megakaryocyte expression. However, megakaryocyte transcription did not fully explain the platelet transcriptome, which had specific pathways systematically enriched or depleted that were obscured by leukocyte contamination. These findings were confirmed in an independent cohort and were associated with protein detection in two platelet proteome resources. In a pilot human sepsis cohort, Siphon-seq reduced transcripts associated with leukocytes to reveal increased platelet-associated pathway signals. These contaminating immune pathways were also enriched in samples analyzed by bulk RNA-seq from an independent sepsis cohort, most strongly in the stratum with the highest estimated leukocyte burden. Combined, these results reveal that RNA-level contamination can alter both transcriptome structure and biological inference, and that Siphon-seq bridges the sensitivity-purity tradeoff that has limited platelet RNA sequencing, enabling sensitive, purity-aware studies of platelet biology in health and disease.
Appeared: Thursday, November 19. bioRxiv. Preprint, not yet peer-reviewed.