Ciliary programmes are inferred, not measured, across the outflow tract endothelial–mesenchymal boundary

Question: How have single-cell and spatial transcriptomic studies resolved endothelial versus mesenchymal ciliary gene expression programmes in the developing cardiac outflow tract, and what module-level analytical frameworks have they used?

Answer

Taken together, the retrieved papers support a deflationary answer. Single-cell and spatial atlases now partition the outflow tract (OFT) cleanly into endocardial/endothelial, neural-crest-derived mesenchymal, second-heart-field (SHF)-derived, and valve-interstitial compartments, and they do so with regulon inference, tissue-niche classification, and spatially coordinated gene-programme detection [5] [4] [24]. None of these frameworks, however, has been applied to a ciliary gene module as such: the module abstractions in use are identity, regulon, niche, and spatial-gradient modules, and ciliary genes are not stratified by compartment anywhere in this corpus. The compartment-resolved cilia answer, insofar as one exists, comes from conditional genetics rather than transcriptomics, and it points the opposite way from a naive expectation: primary cilia are dispensable in endocardium and neural crest but essential in SHF mesenchyme for aortic valve leaflet formation [9]. The field has resolved the endothelial–mesenchymal axis; it has not yet resolved cilia along it.

OFT atlases separate endothelium from mesenchyme by identity and position, leaving ciliary genes unstratified

The human OFT atlas profiles embryonic and fetal OFT plus adult aortic valve derivatives, uses spatial transcriptomics to define the distribution of cell populations, and shows that embryonic signatures persist into adult cells as retrospective lineage labels [5]. The broader developing-heart atlases reach comparable or finer partitions: 63 cell types with location-specific signatures across 4–20 post-conception weeks, including a distinction between coronary and great-vessel endothelial constituents validated spatially [2], and 21 cell types from first-trimester snRNA-seq with Visium deconvolution and novel lymphatic endothelial transcriptomes [8]. These partitions are built from identity markers and location, not from ciliary gene content; no retrieved atlas reports a ciliary module as a discriminating feature between OFT endothelium and OFT mesenchyme.

Regulon and network inference is the dominant module abstraction, and it returns mesenchymal valve programmes rather than ciliary ones

Single-cell regulatory network inference in the OFT identifies GATA6 as a key regulator of valve precursor cells, with a downstream network proposed to contain drivers of human valve defects [5]. Core regulatory network reconstruction in early mesoderm similarly resolves lineage-specific transcription-factor circuits controlling cardiac progenitor specification [1], and GRN reconstruction with functional-redundancy-aware tools such as SyncReg has been used to predict and then validate regulon function by CRISPR Perturb-seq [15]. This framework is powerful for mesenchymal and transitional states — hPSC-derived valvular interstitial cells resolve into two molecularly and functionally distinct subpopulations [17], and single-cell transcriptomics in a Noonan syndrome model reveals widespread dysregulation of extracellular matrix programmes specifically in fibrosa valve interstitial cells [13] — but the regulons recovered are matrix, identity, and lineage modules, not ciliogenesis or intraflagellar transport modules.

Niche and spatially coordinated programme frameworks raise module resolution to tissue scale without reaching organelle-level programmes

Spatial multi-omics of the developing heart defines 19 tissue niches and a context-aware, resolution-agnostic niche classifier (TissueTypist) [4], and integration of prenatal atlases with intact-embryo spatial data charts 114 tissue niches, resolving organ-specific capillary diversification and tissue-specific fibroblast patterning from previously undescribed mesenchymal progenitors [22]. CoPro formalises the same idea for continuously varying programmes, detecting gene programmes that co-vary within or between cell types across space and separating overlapping spatial patterns [24], while SMURF "unrolls" tissue architecture to reveal zonation of expression programmes and the transcription factors regulating them [7]. These are exactly the module-level tools one would want for a ciliary programme comparison; they have simply been pointed at other programmes.

Direct compartment-resolved ciliary evidence comes from genetics, not transcriptomics, and it argues against a strong endocardial ciliary programme

Deleting Ift88 in neural crest cells (Wnt1-Cre) or endocardial-derived cells (Tie2-Cre) did not impair aortic valve formation, whereas deletion in SHF cells (Isl1-Cre) produced bicuspid aortic valve in over half of mutants, via premature SHF differentiation, a broadened intercalated valve swelling, and disrupted Notch–Jag1 signalling [9]. Independent cell-type-resolved cilia mapping outside the heart is consistent with ciliary status being compartment-specific rather than universal: pulmonary endothelial cells lack primary cilia at all developmental stages except in low numbers in proximal pulmonary arteries, while mesenchymal cells are uniformly ciliated early and restricted to PDGFR+ fibroblasts in adult alveoli [3]. Ciliary composition itself is cell-type-specific, with 69% of the ciliary proteome differing between cell lines and 78% of proteins showing single-cilium heterogeneity [12]. IFT proteins additionally carry cilia-independent endothelial functions, with IFT20 controlling VE-cadherin recycling at lymphatic endothelial junctions [20].

Ciliary programme modules are being defined elsewhere with perturbation-anchored designs the OFT field has not imported

A multi-omics comparison of ciliated versus non-ciliated embryonic cells identified SP5 and SP8 as transcription factors upregulated in ciliated cells, and SP8 expression was sufficient to induce primary cilia in unciliated cells [21]. Perturbation-anchored transcriptomics of ciliary ARL13B, integrating independent mutant and rescue clones, produced a high-confidence 131-gene set tracking loss and restoration of ciliary ARL13B, while also demonstrating substantial clone-to-clone transcriptomic heterogeneity [11]. On the endothelial–mesenchymal axis itself, directional EndoMT UP/DOWN gene sets have been curated explicitly to capture partial EndoMT as an active transitional state rather than a mesenchymal endpoint, and benchmarked on bulk and single-cell perturbation data [14]. Network propagation across protein interaction networks has been used to cluster ciliopathies by shared phenotype and prioritise ciliary candidate genes [10]. Each of these is a ready-made module framework; none appears in the OFT atlases retrieved here.

Conflicts and gaps

The main disagreement concerns whether endothelium is ciliated at all. Pulmonary endothelial cells are reported to lack primary cilia throughout development [3], yet endothelial cilia are described during vessel lumen formation, and endothelial-specific ROCK2 ablation causes cilia loss, misorientation, and failed lumen formation with haemorrhage in vivo [19]. Compartment attribution is also coarse: Tie2-Cre labels endocardial-derived cells but the negative Ift88 result does not exclude cilia functioning in a subset of cushion mesenchyme after EndMT [9]. "Mesenchyme" in the OFT is itself heterogeneous — neural crest contributes to OFT mesenchymal populations in assembloid models [6], and cross-species integration shows that mesenchymal continua can be conserved in core but divergent in subpopulations and markers [18] — so a single mesenchymal ciliary module is unlikely to be the right model. Finally, cell–cell communication frameworks that could link an endothelial ciliary signal to a mesenchymal responder are deployed at tissue scale, via topic-modelled interaction patterns [16] or lineage-aware spatial mapping [23], but not for ciliary ligand-receptor logic in the OFT.

What would change the answer

A dedicated ciliary module score — IFT-A, IFT-B, BBSome, transition-zone, and ciliary GPCR gene sets, benchmarked like the ARL13B 131-gene set [11] — applied per cell type in the existing human OFT atlas and its spatial sections would directly test whether endothelium and mesenchyme differ in ciliary programme usage [5]. Pairing that with a targeted Xenium or MERFISH probe panel for ciliary and EndoMT genes, analysed with a spatially coordinated programme detector such as CoPro [24], would distinguish a genuine compartment difference from dissociation artefact. Compartment-specific Ift88 or Kif3a deletion in SHF versus endocardial lineages, read out by snRNA-seq rather than phenotype alone, would convert the current epistasis result into a transcriptional module comparison [9]. Scoring directional EndoMT signatures [14] against ciliary module activity along an endocardial-to-mesenchymal pseudotime would settle whether ciliary programme loss is a cause or a consequence of the transition. Finally, in vivo Perturb-seq of ciliary genes in a patterned cardiac organoid or assembloid system carrying SHF and neural crest contributions [6] would supply the causal module-to-phenotype link that the atlases currently lack.

References

[1] Xie, P., Jiang, X., He, J., Pan, Q., Yang, X., Zheng, Y., Fan, W., Wu, C., Zheng, W., Fang, K., Si, S., Zhu, S., Yang, Y., Zhong, T. P., Yang, Z., Wei, K., Xie, W., Jing, N., Luo, Z., Lin, C.. "Epigenetic delineation of the earliest cardiac lineage segregation by single-cell multi-omics". bioRxiv. 2025-07-16T00:00:00.000Z. URL: https://doi.org/10.1101/2024.05.17.594655

[2] Knight-Schrijver, V. R., Bayraktar, S., Cranley, J., Kanemaru, K., Waller, B., Colzani, M., Wong, C. K., Davaapil, H., Lee, J. C. M., D'Souza, M., Köhne, M. K., Polanski, K., Richardson, L., Semprich, C., Kapuge, R., Dabrowska, M., Mulas, I., Perera, S., Patel, M., Ho, Y., He, X., Tse, H.-F., Tyser, R., Gambardella, L., Teichmann, S. A., Sinha, S.. "A human developing heart atlas reveals cardiovascular cell diversity for in vitro model benchmarking". bioRxiv. 2026-08-06T00:00:00.000Z. URL: https://doi.org/10.1101/2024.04.27.591127

[3] Spurgin, S., Nguimtsop, A. M., Chaudhry, F. N., Michki, S., Salvador, J., Iruela-Arispe, M. L., Zepp, J., Mukhopadhyay, S., Cleaver, O.. "Spatiotemporal dynamics of primary and motile cilia throughout lung development". bioRxiv. 2024-10-26T00:00:00.000Z. URL: https://doi.org/10.1101/2024.10.25.620342

[4] Cranley, J., Kanemaru, K., Bayraktar, S., Knight-Schrijver, V., Pett, J. P., Polanski, K., Dabrowska, M., Hulbert, R., Lana-Elola, E., Aoidi, R., Mulas, I., Richardson, L., Semprich, C., Kapuge, R., Perera, S., He, X., Ho, S. Y., Yayon, N., Tuck, L., Roberts, K., Palmer, J., Davaapil, H., Gambardella, L., Patel, M., Tyser, R., Tybulewicz, V., Sinha, S., Teichmann, S.. "Developmental Dynamics of Human Cardiogenesis: A multi-omic reference and its disruption in Trisomy 21". bioRxiv. 2025-08-28T00:00:00.000Z. URL: https://doi.org/10.1101/2024.04.29.591736

[5] Leshem, R., Murtuza Baker, S., Mallen, J., Wang, L., Dark, J., Rattray, M., Sharrocks, A., Piper Hanley, K., Hanley, N., Bamforth, S., Bobola, N.. "A cell atlas of the developing human outflow tract of the heart and its adult aortic valve derivatives". bioRxiv. 2026-03-03T00:00:00.000Z. URL: https://doi.org/10.1101/2023.04.05.535627

[6] Kostina, A., Kiselev, A., Huang, A., Lankerd, H., Caywood, S., Jurado-Fernandez, A., Volmert, B., O'Hern, C., Juhong, A., Liu, Y., Qiu, Z., Park, S., Aguirre, A.. "Self-organizing human heart assembloids with autologous and developmentally relevant cardiac neural crest-derived tissues". bioRxiv. 2024-12-11T00:00:00.000Z. URL: https://doi.org/10.1101/2024.12.11.627627

[7] Guo, J., Sarafinovska, S., Hagenson, R., Valentine, M., Chen, D. Y., McCoy, W. H., Dougherty, J., Mitra, R. D., Muegge, B. D.. "SMURF: soft-segmentation for single-cell reconstruction and topological analysis of spatial transcriptomic data". bioRxiv. 2025-11-01T00:00:00.000Z. URL: https://doi.org/10.1101/2025.05.28.656357

[8] De Bono, C., Xu, Y., Kausar, S., Herbane, M., Humbert, C., Rafatov, S., Missirian, C., Moreno, M., Shi, W., Gitton, Y. I., Lombardini, A., Vanzetta, I., Mazaud-Guittot, S., Chedotal, A., Baudot, A., Zaffran, S., Etchevers, H. C.. "Multi-modal refinement of the human heart atlas during the first gestational trimester". bioRxiv. 2025-01-29T00:00:00.000Z. URL: https://doi.org/10.1101/2024.11.21.624698

[9] Alqahtani, A., Eley, L., Newton, J., Macdonald, K., Connolly, C., Rodrigues-Cleto, C., Neffeova, K., Lopez, L., Arias, J., Derrick, C. J., Alaradi, M., Kolesova, H., Chaudhry, B., Henderson, D. J.. "Remodelling of supernumerary leaflet primordia leads to bicuspid aortic valve (BAV) caused by loss of primary cilia". bioRxiv. 2025-02-25T00:00:00.000Z. URL: https://doi.org/10.1101/2025.02.25.636485

[10] Aarts, E. M., Laman Trip, D. S., Neatu, R., Martin, C. G., Riley, B., Kraus, A., Green, A., Al-Hamed, M. H., Armstrong, R. E., Sayer, J., Bachmann-Gagescu, R., Beltrao, P.. "Network-based framework for studying etiology and phenotype diversity in primary ciliopathies". bioRxiv. 2025-01-09T00:00:00.000Z. URL: https://doi.org/10.1101/2025.01.08.631887

[11] Morrison, O., Caspary, T.. "Transcriptomics of independent CRISPR-edited cell lines reveal ciliary-specific ARL13B dependent changes". bioRxiv. 2026-10-08T00:00:00.000Z. URL: https://doi.org/10.64898/2026.08.13.744725

[12] Hansen, J. N., Sun, H., Kahnert, K., Westenius, E., Johannesson, A., Villegas, C., Le, T., Tzavlaki, K., Winsnes, C., Pohjahnen, E., Maekiniemi, A., Fall, J., Navarro, F. B., Backstrom, A., Lindskog, C., Johansson, F., von Feilitzen, K., Vega, A. D., Casals, A. M., Mahdessian, D., Uhlen, M., Sheu, S.-H., Lindstrand, A., Axelsson, U., Lundberg, E.. "Intrinsic Heterogeneity of Primary Cilia Revealed Through Spatial Proteomics". bioRxiv. 2025-05-23T00:00:00.000Z. URL: https://doi.org/10.1101/2024.10.20.619273

[13] Liu, C. Z., Patel, S., Sidoli, S., Prasad, A., Charytonowicz, D., Mazine, A., Mikryukov, A. A., Abdul-Ghafar, J., Kahn, E. S., Young, D., Porter, G. A., Katzman, P. J., Sanders, S. P., Carreon, C. K., Hubmacher, D., Lincoln, J., Keller, G., Chung, W. K., Sebra, R., Gelb, B. D.. "Dysregulated TGFb-ERK Signaling Drives Aberrant Extracellular Matrix Production in Noonan Syndrome-Associated Pulmonary Valve Stenosis". bioRxiv. 2026-01-20T00:00:00.000Z. URL: https://doi.org/10.64898/2026.01.16.700032

[14] Best, K., Casement, J., Lovat, P. E., Walker, L., Armstrong, J.. "EndoMT: An R Package Resource for Endothelial-Mesenchymal Transition Gene Set Enrichment Analysis". bioRxiv. 2026-10-05T00:00:00.000Z. URL: https://doi.org/10.64898/2026.09.29.755303

[15] Hu, Z., Mayes, S., Wang, W., Santos-Pereira, J. M., Theis, F., Sauka-Spengler, T.. "Endothelial regulatory circuits control cranial neural crest migration and plasticity". bioRxiv. 2025-02-27T00:00:00.000Z. URL: https://doi.org/10.1101/2024.09.17.613303

[16] Zhu, Z., Zheng, R., Yu, Y., Zhang, L., Chen, K.. "Human single-cell atlas analysis reveals heterogeneous endothelial signaling". bioRxiv. 2025-09-27T00:00:00.000Z. URL: https://doi.org/10.1101/2025.08.15.670370

[17] Mazine, A., Mikryukov, A. A., Fernandes, I., Liu, C. Z., Jahangiri, S., Martin, M., Gähwiler, E. K., Generali, M., Gomez, J., Latifi, N., Miao, Y., Liu, Y., Laflamme, M., Simmons, C. A., Hoerstrup, S. P., Emmert, M. Y., Gelb, B. D., Gu, M., Keller, G.. "Generation of Valvular Interstitial Cells from Human Pluripotent Stem Cells". bioRxiv. 2025-09-21T00:00:00.000Z. URL: https://doi.org/10.1101/2025.09.17.676963

[18] Sharysh, D., Gomez-Imaz, P. N., Cano, D. M., Markov, A., Serrano, R. I., Fitera, L. C., Albarran-Juarez, J., Bentzon, J. F.. "Mouse, pig, and human atherosclerotic lesions have common and distinct mesenchymal cell populations". bioRxiv. 2025-08-02T00:00:00.000Z. URL: https://doi.org/10.1101/2025.07.24.666663

[19] Mavria, G., Zahed Mohajerani, S., Grant, G., Mccarthy, A., Bourn, M. D., Peyman, S. A., Johnson, C. A.. "Microfluidic analysis reveals ROCK2 regulation of endothelial cilia is essential for blood vessel lumen formation and vascular integrity". bioRxiv. 2026-05-28T00:00:00.000Z. URL: https://doi.org/10.64898/2026.05.27.728336

[20] Paulson, D., Majumder, A., Lehmann, Z., Knutson, L., Paulson, J., Lasey, S., Fink, D. M.. "IFT20 regulates lymphatic endothelial cell-cell junctions via endocytic trafficking of VE-cadherin". bioRxiv. 2025-01-15T00:00:00.000Z. URL: https://doi.org/10.1101/2025.01.15.631989

[21] Liang, Y., Koche, R., Chalamalasetty, R. B., Stephen, D. N., Kennedy, M. W., Lao, Z., Pang, Y., Kuo, Y.-Y., Lee, M., Lobo, F. P., Huang, X., Hadjantonakis, A.-K., Yamaguchi, T. P., Anderson, K. V., Joyner, A. L.. "Transcription factors SP5 and SP8 drive primary cilia formation". bioRxiv. 2025-06-03T00:00:00.000Z. URL: https://doi.org/10.1101/2025.06.03.657415

[22] Webb, S., Rose, A., Xu, C., Steele, L., Kuri, M. A., Stephenson, E., Inecik, K., Jafree, D., Foster, A. R., Basurto-Lozada, D., Chipampe, N.-J., Pournara, A. V., Jacques, M.-A., To, K., Admane, C., Kritikaki, E., Chroscik, M. A., Horsfall, D., Foreman, J., Rademaker, K., Karjalainen, J., Laddach, A., Madad, S., Lawrence, J. E. G., Kleshchevnikov, V., Lisgo, S., Lee, J. T. H., Blevinal, J., Alqahtani, A., Makarchuk, S., Jackson, J., Ucuncu, E., Silva, T. P., Lorenzi, V., Torabi, F., Botting, R. A., Roberts, K., Olabi, B., Chakala, K. P., Dony, L., Dall'Aglio, G., Cujba, A.-M., Whitfield, H. J., . "An integrated single cell and spatial omics atlas of human prenatal development". bioRxiv. 2026-04-21T00:00:00.000Z. URL: https://doi.org/10.64898/2026.03.30.714220

[23] Pan, X., Danies-Lopez, A., Zhang, X.. "Mapping lineage-resolved scRNA-seq data with spatialtranscriptomics using TemSOMap". bioRxiv. 2026-01-20T00:00:00.000Z. URL: https://doi.org/10.1101/2024.10.31.621331

[24] Miao, Z., Qu, Y., Huang, S., Laux, L., Peters, S., Aristel, A., Zhang, Z., Niedernhofer, L. J., McMahon, A., Kim, J., Zhang, N.. "Dissecting the coordinated progression of cell states in spatial transcriptomics with CoPro". bioRxiv. 2026-04-21T00:00:00.000Z. URL: https://doi.org/10.64898/2026.04.17.719309

How this survey was made

This survey answers one question: “How have single-cell and spatial transcriptomic studies resolved endothelial versus mesenchymal ciliary gene expression programmes in the developing cardiac outflow tract, and what module-level analytical frameworks have they used?”

We keep our own collection of recent preprints and refresh it every day from bioRxiv, medRxiv, arXiv, PsyArXiv, SocArXiv and ChemRxiv. For each survey, we search that collection for recent work and keep the strongest matches: papers that use the same terms as the question, and papers on the same topic in different words.

The text was written from the titles and abstracts of those papers; the full texts were not read. Statements that draw on a particular paper are marked with a number in brackets. The reference list just above is the 24 papers whose work this text draws on, chosen from the 129 papers those searches returned; the rest of the shortlist is not printed. It is not a complete review of everything published on the question.

Treat this as a starting point and a reading list, not a peer-reviewed review. The searches behind this survey ran on October 22, 2026 and covered work from the previous two years.