Hackathon Challenges

Data QC / Visualisation / Setting the benchmark for an API

Topics pertaining to performing quality control on the data, visualising clones, and developing the APIs to support this.

This includes:

• Developing the appropriate data structure to deal with large volume of single-cell clonality data.
• Extending Bioconductor’s SingleCellExperiment S4 class in R to include a slot for storing barcode information.
• Extending the concept of meta-cell in scRNAseq to meta-clone.
• Developing APIs to perform data QC and to assess the results.
• Developing an API to better visualise expanded clones and pedigrees in scRNAseq datasets.
• Visualisation of pedigrees and framework for annotation (div number, state, etc)

Differential expression within and between clones

Topics pertaining to performing differential expression and abundance analysis between the clones.

This includes:

• Quantifying intra- and inter-clonal transcriptional signatures.
• Can we quantify clonal expansion and proliferation capacity?
• Determining how to best quantify the distribution of clonally related cells across different cell states / tissues.
• Determining how to show clonal relationships while respecting GEX similarity, or TCR/BCR similarity at a scalable fashion.

Identify clonally related cells based on transcriptome similarity

Topics pertaining identifying clonally related cells based on their transcriptomic profile (independent of their barcodes).

This includes:

• Barcode free detection of clones i.e. are there genes that are unrelated to cell state, but heritable clonally?
• Overcoming the limits of sampling - how to deal with missing clonal members.
• How to deal with “soup” V(D)J contigs?
• Are there better ways to understand the molecular determinants of clonal fate beyond linear regression?