Jan Lause

Hey Thorsten, thanks for your quick reply, it was very helpful! Based on your advice, I wrote a little python function that runs 2D segmentation tasks using your caffe_unet and...

Happy to hear that ;) I will maintain/update the code here: https://github.com/jlause/unet-segmentation-python

PS: Minimal example: ``` adata = sc.datasets.pbmc3k() sc.pp.filter_genes(adata,min_cells=1) print(adata.shape) output = sc.pp.highly_variable_genes(adata,flavor='cell_ranger',inplace=False,subset=True) print(adata.shape) print(output) ```

> The slack export viewer was probably designed for official Slack exports, not slackdump-like. therefor it is not able to reference local files. Flask (the internal webserver used here) is...

PS: Did not work first try - e.g. I have an image link in the slackviewer html `http://localhost:5000/channel/channelname/attachments/F03NMKELY03-image.png` and made a folder `/home/user/miniconda3/envs/myenv/lib/python3.10/site-packages/slackviewer/static/channel` into which I copied the exported `channelname`...

Hi! I‘ve implemented the Pearson residuals in scanpy, but I don’t know scrublet so well. After quickly glancing over the paper I thought that in principle Scrublet should be able...

Hey all, thanks for all this input! Regarding @LuckyMD's point of the 'one-stop-preprocessing': I like the idea of having a complete workflow like `HVG selection -> transform counts to pearson...

Regarding the practical implementation of `sc.pp.normalize_pearson_res(adata)`, @dkobak proposed to integrate usage of previously computed gene selections: > This function could have an optional argument `use_highly_variable=False` and if it's set to...

@dkobak @LuckyMD Regarding `sc.pp.recipe_pearson_residuals()` vs. `sc.pp.normalize_pca_pearson_residuals()`: First of all: I thought having the pipeline as a recipe might be appropriate because it would ship the 'standard approach' we suggest in...

Super, danke fuer die schnelle Hilfe! Das hilft mir, dann kann doch ich via lokalen Webserver die config.json bearbeiten. Danke vielmals und schoenes Wochenende!