Tomasz Konopka

Thanks for merging @plouc and also for the review earlier. If any issues come up, I'm happy to help edit and improve this further - just let me know

The underlying bug has been fixed in d3-shape v3.2, so this patch is no longer needed.

Thanks for reporting. That config seems to match the default (a=NA, b=NA, spread=1) so am not sure why you are seeing the warning. Could you confirm that is indeed the...

Thanks for the detailed investigation. Glad you found the root cause and can work around this for now. As you suggest, additional input validation would be useful. Most inputs should...

Encountered similar issues when drawing short chromosomes and also near chromosome edges. I made some edits in a branch in a fork. ([branch cli_margin](https://github.com/tkonopka/bamsnap/tree/cli_margin)) An example synthetic fasta sequence and...

Thanks for asking. Yes, citations should go to the original authors. The umap-learn repository lists recommendations [here](https://github.com/lmcinnes/umap#citation). In brief, you can cite the [manuscript on arxiv](https://arxiv.org/abs/1802.03426) or the [paper in...

Hi Marta. Thanks for using the package and reaching out. Yes, your understanding is correct: setting the metric to "pearson" or to "cosine" will give the same results. The reasons...

Yes, centering will make the distances consistent with interpretation as pearson and pearson-squared. Btw, since you're already deep in the code, the implementation of the distances are in file `src/distances.Rcpp`....

Hi Jayani - unfortunately the loss function value doesn't get propagated to the output object. But there is access to the embedding (of course) and also the knn details, so...