Gregor Sturm

spatial airr data

5

### Description of feature Apparently this is a thing now (e.g. 10x spatial transcriptomics + mixcr, e.g. in [this paper](https://www.sciencedirect.com/science/article/pii/S1074761322000814?via%3Dihub)). @giovp is there anything we need to do to leverage...

Speed up ir_dist

2

Parasail is highly optimized - but it is invoked for every sequence individually from Python which creates a bottleneck. We could solve this issue by integrating with external tools that...

Consistent typing in adata.obs

3

* Most columns added to `adata.obs` should probably be well-defined categoricals * Booleans should be booleans / or categoricals, and not strings * `nan` should be `nan` and not `"nan"`....

Add more epitope databases to `datasets`

3

List adapted from #54. - [ ] The [MAIT reference database](http://www.cbs.dtu.dk/services/MAIT_Match/MAIT_DB_update.php) with invariant CDR3 - [ ] huARdb https://academic.oup.com/nar/advance-article/doi/10.1093/nar/gkab857/6381136?rss=1 - [ ] McPAS-TCR - [ ] PIRD - [ ]...

airr validation fails when converting demo dataset to_airr_cells

1

### Description of the bug Can't convert demo dataset `to_airr_cells` because of AIRR validation error. ### Minimal reproducible example ```python import scirpy as ir adata = ir.datasets.wu2020_3k() ir.io.to_airr_cells(adata) ``` ###...

improve docs build system

4

*Current list of stuff to solve*: - [x] build an overview page of available versions (see also #267) - [ ] move docs to separate repo to keep footprint small...

Improve clonotype network plot with plotAtlas

1

Networkx now offers an "Atlas" example which could be useful for making nicer, packed plots that the current approach using "squarify". Not sure how it handles node sizes and probably...

Refactoring group_abundance.

2

The current group_abundance function is relatively slow. Also I don't like the name too much. barplot or stackedbarplot is better. `x`, `y` and `color` are better argument names. Maybe this...

In tl.clonal_expansion replace `clip_at` with `breakpoints`

1

For datasets with higher clonal expansion, it would be great to define custom breakpoints (e.g. 1,2,5,10,20,50) for clonal expansion rather than the 1,2,3,...,>=n that is currently possible with `clip_at`.

It would be nice being able to run [conga](https://github.com/phbradley/conga) from scirpy. A wrapper like `scirpy.external.tl.conga` could work. See also https://github.com/phbradley/conga/issues/17 for some discussion.