Bioinformatics Research Engineer, involved in the following projects:
- Development of a method to get the transcriptome of single cells at two different timepoints.
This project relies on scNMT-seq and uses zebrafish as model organism. By coupling a methyltransferase to the polymerase-II at time t0, only active genes get methylated. We then compare the methylation patterns at time t1 to the RNA-seq at time t0 to get the transcriptome “from the past”.
Tools at use: Seurat, snakemake, sequana for demultiplexing, STAR alignment, BISCUIT alignment, custom scripts for extensive analysis of the zebrafish genome to check for biases in the methyltransferase binding sites, UCSC genome browser custom tracks. - Visium HD single cell spatial transcriptomics:
to identify heterogeneity in craniofacial muscles of human embryo. This project focuses on 4 slides of human embryo head: two sagittal sections of head and two sections of extraocular muscles (EOM). Using visium HD, combined with DeepLearning segmentation, we properly identified markers of development in the Tbx1-independent branch of craniofacial development. By differential expression analysis between the EOM and other Tbx1-independent muscles, we identify candidate markers and then visualize their gradient of expression at two different developmental stages using pseudotime analysis of scanpy. The data is integrated with a scRNA-seq atlas of human embryo at various stages of development to valide the newly identified markers of mesodermal muscle development.
Tools at use: Seurat, Scanpy, StarDist, sopa, SpatialData, AnnData.
I also have expertise in Machine Learning which I am keen to apply to various projects.
Ongoing:
- Xenium Spatial Transcriptomics
- Integration of visium HD with scRNA-seq data