Zureick, N. | Search bioRxiv

bioRxiv2026

CMAC: A deep learning framework for absolute cardiomyocyte transcriptional age estimation

Cardiomyocyte maturation spans embryogenesis through postnatal life and involves coordinated transcriptional transitions that are disrupted in disease and incompletely recapitulated by stem cell-derived cardiomyocytes. However, no unified framework exists for assigning absolute transcriptional maturation age across datasets, platforms, and experimental conditions. Here, we developed CMAC (Cardiomyocyte Maturation Age Clock), integrating 18 murine single-cell RNA-sequencing datasets spanning embryonic and postnatal timepoints, into a batch-invariant probabilistic atlas using single cell variational inference (scVI) with a 100-dimensional latent representation. Analysis of the developmental trajectory revealed highly non-uniform transcriptional change, with 63.9% of the total transcriptional journey completed by birth and the neonatal transition representing the largest single transcriptional step. An XGBoost regressor trained on the CMAC latent space predicted absolute chronological age with a leave-one-dataset-out mean absolute error (MAE) of 4.09 days (median 3.27 days), Pearson r = 0.903, and Spearman {rho} = 0.908 across 18 datasets and six sequencing platforms, outperforming linear regression (MAE = 6.18 days) and PCA-based representations (MAE = 6.35 days). In an independent P14 cohort comprising 98,163 cardiomyocytes from six biological replicates, CMAC achieved a mouse-level MAE of 0.33 days, with all six mice predicted within 1.1 days of chronological age. Application of CMAC to a previously published PGC1/{beta} double-knockout model with established delayed cardiomyocyte maturation independently recapitulated this phenotype, quantifying maturation deficits of 4.77 to 8.60 days across P7 to P28, with the largest deficit at P14. Together, these results establish CMAC as a quantitative, deep learning framework for assigning absolute transcriptional maturation age and quantifying deviations from normal cardiomyocyte development.

genomics↗

bioRxiv2026

Splicing Factor SF3B2 Regulates Cardiomyocyte Calcium Handling Through Alternative Splicing of Cardiac Ion Channel Genes

BackgroundAlternative splicing is a critical determinant of protein diversity in the heart, where it drives the postnatal functional maturation of cardiomyocytes and specifies the ion-channel and calcium-handling isoforms required for mature contractile function; dysregulated splicing programs have in turn been implicated in cardiomyopathies and arrhythmias. However, the splicing regulators that control cardiomyocyte calcium handling remain largely unknown. ObjectiveWe systematically screened 276 splicing factor genes in human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) to identify regulators of calcium handling, and we selected SF3B2, a core U2 snRNP spliceosome component, for detailed follow-up based on complex-level enrichment and prior identification by Murphy et al. 2021 [1]. MethodsA high-throughput siRNA screen targeting 276 splicing factor genes was performed in hiPSC-CMs using 384-well calcium transient imaging. SF3B2 knockdown was followed by deep bulk RNA-seq analysis (n = 3 per group) for differential gene expression with DESeq2 and alternative splicing quantification with rMATS. RNA immunoprecipitation sequencing (RIP-seq) using an epitope-tagged SF3B2 construct was performed to identify direct mRNA binding targets of SF3B2. ResultsThe screen identified multiple U2 snRNP components, including SF3A2, SF3B1, and SF3B4--as regulators of calcium transient duration. SF3B2 was previously identified as a regulator of contractility in this screen [1], and the enrichment of its complex partners above the significance threshold in the current CTD75 analysis supported its selection for follow-up characterization. SF3B2 knockdown resulted in 2,683 differentially expressed genes (adjusted p < 0.05), with downregulated genes enriched in cell cycle and DNA replication pathways. Alternative splicing analysis revealed significant changes across all five rMATS event types in 36 genes within the cardiac muscle cell action potential involved in contraction gene ontology term (GO:0086002), including calcium channel (CACNA1C, CACNA1D, CACNA2D1), potassium channel (KCNH2, KCNQ1), and sodium channel (SCN5A) genes, although the large number of affected genes is consistent with broad spliceosomal perturbation and a formal enrichment test would be needed to determine whether cardiac action potential genes are preferentially affected. Integration of RIP-seq data (597 SF3B2-enriched transcripts, FDR < 0.05) with splicing analysis identified CACNA2D1, which encodes an auxiliary subunit of L-type calcium channels, as both directly bound by SF3B2 and alternatively spliced upon knockdown. ConclusionThese findings identify SF3B2 as a regulator of cardiomyocyte calcium handling and suggest that SF3B2-dependent missplicing of CACNA2D1 may link core spliceosome function to the splicing programs underlying cardiomyocyte functional maturation. Significance StatementThis study provides the first systematic functional screen of splicing factors in cardiomyocyte calcium handling and identifies SF3B2, a U2 snRNP subunit, as a regulator of cardiac ion channel splicing. Transcriptomic, splicing, and RNA binding data converge on CACNA2D1, identifying a route by which a core splicing factor shapes cardiac contractility, extending cardiac splicing regulation beyond the accessory RBPs studied to date.

developmental biology↗