README.md

In jgilis/iSEEtranscripts: Unlocking iSEE for transcript-level visualization

Goal

To unlock to the iSEE bioconductor package for transcript-level visualization in the context single-cell omics experiments. The idea is to have additional panels to supplement the established (gene-level) visualizations of iSEE with relevant transcript-level plots. An end-user should be able to select a gene or transcript of interest and to readily obtain relevant information at the level of the individual isoforms.

Specific challenges

• The first obvious challenge is to come up with transcript-level visualization strategies that have an added value for end-users. Deciding on which plots are most relevant is a work in progress, but the following types have been put forward at the moment.

Visualizing the raw usage of transcript, i.e. by generating a violin plot on the fraction of the expression of a target transcript and the expression of its corresponding gene:

Adopting/adapting the DTU visualization provided by DRIMSeq:

Visualizing gene models as provided in ggbio. Note that this requires interaction with GRanges, which has not been implemented in iSEE yet.

• One of the main conceptual challenges will be to make the iSEEtranscripts project (and the iSEE project as a whole) as flexible as possible. Therefore, we should consider which input type is optimally suited to assure flexibility. For instance should we work with single (Ranged-)SummarizedExperiments where the transcript-level counts and gene-level counts are separate assays, or should we leverage altExp or MultiAssayExperiment to facilitate the visualization of multimodal data with iSEE.

Structure

This project should obviously be compatible with existing bioconductor concepts, including established data types and workflows. Therefore, we suggest the following workflow during the early stages of project;

1. Work on a single reference dataset; macrophage
2. Import data using tximport and/or tximeta.
3. Wrangle data in the desired object type (see specific challenges)
4. Explore different visualization strategies (see specific challenges)

Website

Link to pkgdown website: https://jgilis.github.io/iSEEtranscripts/

Collaborate

In order to make this a collaborative effort, we suggest forking the original iSEEtranscripts repository, working on a branch of your fork, and submitting a pull request when your contribution is ready for review. In addition, we will use the bioc slack channel eurobioc2020-isee-transcripts for further communication and brainstorming.

BiocChallenges

This project is created in light of the BiocChallenges project introduced at eurobioc2020.

jgilis/iSEEtranscripts documentation built on Aug. 14, 2022, 3:35 a.m.