This repository contains all the code and computational work undertaken as part of my (RP1) dissertation submitted to The University of Manchester for the degree of Master of Science in Bioinformatics and Systems Biology in the Faculty of Biology, Medicine and Health.
The placenta is the organ with the greatest inter-species diversity in morphological and histological features. Great apes, including humans, undergo haemochorial placentation characterised by deep trophoblast invasion to establish direct contact with the maternal blood supply in the endometrium. The deepest trophoblast invasion has been observed in humans, and it is considered an evolutionary progression driving the development of the human brain, the largest among the great ape's brains [1-3]. However, this evolutionary adaptation combined with bipedalism resulted in more frequent human-specific pregnancy complications [2]. Despite that, interactome networks mediating human trophoblast invasion and placentation have not been well explored. This project analysed single-cell transcriptomic data from the first-trimester placenta and blastocyst, followed by the hypergraph model development. The hypergraph output was mapped to evolutionary constraint metric, LOEUF score, and human accelerated regions (HARs) to identify genomic markers of evolutionary impact in the early human placenta. The results identified 932 evolutionarily constrained genes and 16 genes overlapping with HARs. Further functional analysis identified molecular bridges between placental development, gastrulation, trophoblast cell migration, cell-matrix adhesion and epithelial-mesenchymal transition, including EOMES, FBN2, and ITGA5. This project delivered a functional hypergraph model able to identify genomic markers of recent human evolution in complex biological systems and molecular bridges connecting different biological processes involved in both placentation and tumourigenesis.
[1] Burton, G. J., & Jauniaux, E. (2023). The human placenta: new perspectives on its formation and function during early pregnancy. Proceedings of the Royal Society B: Biological Sciences, 290(1997), 20230191. https://doi.org/10.1098/rspb.2023.0191
[2] Burton, G. J., & Fowden, A. L. (2015). The placenta: a multifaceted, transient organ. Philosophical Transactions of the Royal Society B: Biological Sciences, 370(1663), 20140066. https://doi.org/10.1098/rstb.2014.0066
[3] Abbas, Y., Turco, M. Y., Burton, G. J., & Moffett, A. (2020). Investigation of human trophoblast invasion in vitro. Human Reproduction Update, 26(4), 501–513. https://doi.org/10.1093/humupd/dmaa017
The project repository is structured in the following manner:
001_RP1/
├── single-cell-analysis/
│ ├── code/
│ ├── data/
│ │ ├── combat_xiang_v_others_filter200_with_monocle.h5ad
│ │ ├── MSc_processed_vento_tormo_data.h5ad
│ │ └── tidy_data/
│ │ ├── blastocyst_data.h5ad
│ │ ├── first_trimester_placental_data.h5ad
│ │ ├── integrated_data.h5ad
│ │ └── integrated_data_normalized.h5ad
│ ├── output/
│ │ ├── adata_obs.csv
│ │ ├── adata_var.csv
│ │ ├── adata_X.csv
│ │ ├── DEGs_blastocyst/
│ │ ├── DEGs_integrated/
│ │ ├── DEGs_placenta/
│ │ └── visualisations/
│ └── training/
│ └── 2025-02-15-scRNA-seq-training.ipynb
│
└── hypergraph-analysis/
├── R-analysis.Rproj
├── code/
│ ├── notebooks/
│ ├── scripts/
│ └── host-lab-scripts/
├── data/
│ ├── adata_obs.csv
│ ├── adata_var.csv
│ ├── adata_X.csv
│ ├── glycosmos_homosapiens_glycogenes.csv
│ ├── glycosmos_homosapiens_glycogenes.xlsx
│ ├── HARs/
│ │ ├── raw/
│ │ ├── science.abm1696_table_s1.xlsx
│ │ └── science.abm1696_table_s2.xlsx
│ └── loeuf/
│ └── gnomad.v4.1.constraint_metrics.tsv
└── output/
├── all-ranked-constrained-and-HAR-mapped-genes.csv
├── all-rank-LOEUF-mapped-genes.csv
├── filtered-ranked-constrained-and-HAR-mapped-genes.csv
├── filtered-ranked-constrained-and-HAR-mapped-genes-summary.csv
├── filtered-rank-LOEUF-mapped-genes.csv
├── ranked_glycogenes_fixed.csv
├── har_genes_info.csv
├── cell_annotations.csv
├── cnetplot_key.png
├── bio_process_heatmap_functional_classification.png
├── GO/ (28 files: .csv, .png)
├── KEGG/ (16 files: .csv, .png, .xml)
└── hypergraph_analysis/
├── adjacency_matrix/
├── incidence_matrix/
├── plots/
└── hypergraph_results_spearman_*.csv
Note: data/ and output/ are gitignored in this repository. The supply of raw and intermediate data files to reproduce the analysis is at the discretion of the host lab.
This project combines a Python workflow (single-cell-analysis/) and an R workflow (hypergraph-analysis/). Package requirements and versions are listed below.
| Package | Version |
|---|---|
| Python | 3.11.8 |
| Jupyter Lab | 4.3.5 |
| scanpy | 1.10.4 |
| anndata | 0.11.3 |
| pandas | 2.2.2 |
| numpy | 2.1.3 |
| matplotlib | 3.10.0 |
| janitor | 0.31.0 |
| Package | Version | Notebooks |
|---|---|---|
| R | 4.4.1 | All |
| RStudio | 2024.12.1+563 | All |
| tidyverse | 2.0.0 | All |
| rstatix | 0.7.2 | 2025-01-10-hypernetwork-analysis.Rmd, 2025-04-26-statistical-testing.Rmd |
| dendextend | 1.19.1 | automated-hypernetwork-analysis.Rmd |
| gplots | 3.2.0 | automated-hypernetwork-analysis.Rmd |
| plyr | 1.8.9 | automated-hypernetwork-analysis.Rmd |
| ggpubr | 0.6.1 | automated-hypernetwork-analysis.Rmd, 2025-04-15-loeuf-mapping.Rmd |
| ggsci | 3.2.0 | automated-hypernetwork-analysis.Rmd, 2025-04-15-loeuf-mapping.Rmd |
| BioQC | 1.34.0 | automated-hypernetwork-analysis.Rmd |
| readxl | 1.4.5 | 2025-04-17-HAR-mapping.Rmd |
| GenomicRanges | 1.58.0 | 2025-04-17-HAR-mapping.Rmd |
| GenomicFeatures | 1.58.0 | 2025-04-17-HAR-mapping.Rmd |
| AnnotationDbi | 1.68.0 | 2025-04-17-HAR-mapping.Rmd |
| TxDb.Hsapiens.UCSC.hg38.knownGene | 3.20.0 | 2025-04-17-HAR-mapping.Rmd |
| org.Hs.eg.db | 3.20.0 | 2025-04-17-HAR-mapping.Rmd |
| annotables | 0.2.0 | 2025-04-17-HAR-mapping.Rmd |
| clusterProfiler | 4.14.6 | 2025-04-17-HAR-mapping.Rmd |
| enrichplot | 1.26.6 | 2025-04-17-HAR-mapping.Rmd |
| pathview | 1.46.0 | 2025-04-17-HAR-mapping.Rmd |
| janitor | 2.2.1 | ranked_glycogenes.Rmd |
The part of this work in ranked_glycogenes.Rmd contributed to the following conference abstract publication:
Chukwu, J., Marčec, D., Jones, C., Aplin, J., & Stevens, A. (2025). Modelling the function of glycosylation genes expressed in early placental development using human single-cell RNAseq data and evidence for evolutionary impact in placental mammals and hominids. Placenta. https://doi.org/10.1016/j.placenta.2025.08.194
If you use or refer to this work, please cite it as:
Marčec, D. (2026). Identifying genomic markers of evolutionary impact on the early human placenta [Master's thesis, University of Manchester]. GitHub. https://github.com/doramarcec/RP1
Citation metadata is also available in CITATION.cff (used by GitHub's "Cite this repository" feature) for BibTeX and other formats.
The code in this repository is licensed under the MIT License. The dissertation text, abstract, and figures are made available under CC-BY-4.0.