Improve visualization functions

[OMIC-coding]

In the key figures from the paper, such as Fig. 2e and Fig. 3c, there are no readily available functions to directly reproduce the visualizations. Additionally, in the default HTML output, spots not assigned to any component are displayed in black rather than the gray used in the paper, which makes it difficult to distinguish key information—particularly for Visium HD data. It would be greatly helpful to provide built-in visualization functions for these figures and to adjust the default color scheme for unassigned spots to improve interpretability.

[anne04]

Thank you for the question, and sorry for the inconvenience.

1. Fig 2e: Can be produced by following this vignette: https://github.com/schwartzlab-methods/CellNEST/blob/main/vignette/filter_ccc_list_for_type_region.md#filter-by-ligand-receptor-pair:

Basically, varying the number --top_edge_count=10000/5000/500 ... in the following command (for details, please see that vignette):

cellnest visualize --data_name='V1_Human_Lymph_Node_spatial' --model_name='CellNEST_V1_Human_Lymph_Node_spatial' --top_edge_count=10000 --filter=1 --filter_by_ligand_receptor='CCL19-CCR7'

2. Fig. 3C: can be produced by: https://github.com/schwartzlab-methods/CellNEST_paper_figures/blob/main/Human_lymph_TCellZone_relay_plot.py
   However, we agree this script is not well-documented, and we are working to make it easy to access.

3. It would be greatly helpful to provide built-in visualization functions for these figures and to adjust the default color scheme for unassigned spots to improve interpretability --- Thank you very much for the suggestion. We will add this functionality soon and let you know here.

Thanks.

[OMIC-coding]

Thank you for the detailed guidance and links. However, I still believe it would be highly beneficial to integrate these plots as built-in functions, since they are rather fundamental visualizations rather than advanced analyses. This would greatly improve user experience and lower the entry barrier, especially for new users.

I have also been actively promoting CellNEST on WeChat, a major Chinese academic communication platform, which could help increase citations and visibility. Many potential users are interested, but a common piece of feedback is that the current visualization options are limited and sometimes unclear, so improving this aspect would be a significant boost for adoption.

In addition, while testing CellNEST across different disease contexts and tissue types, I noticed that certain genes such as TGFB1 and LRP1/2 consistently rank among the top results. This pattern appears in diverse datasets, which makes me wonder whether this is fully biologically meaningful or may be due to some model bias or general connectivity patterns. Could you please share your thoughts on how to interpret such recurring top-ranked genes?

Thanks again for your work on this great tool, and I look forward to future improvements.

[Sawakotx]

Same with him, looking forward to optimized visualization!!!

[OMIC-coding]

Thank you for the clarification and the vignette link. I tried running the command you suggested for Fig 2e, but I did not see any figure output. Instead, the command only generated a .csv file as output. Could you please let me know if there is an additional argument or step required to produce the figure?

[BingqiXu]

I'm really interested in and impressed by this cell communication algorithm - it's a fascinating approach!
Unfortunately, I've encountered the same issue mentioned above while working with the implementation.
Beyond that, I also wanted to share that I think enhancing the visualization components would greatly improve the usability of this tool. It would help users better interpret and present the results of the cell communication analysis.
Would you be open to providing some guidance on resolving the issue, and possibly considering improvements to the visualization features in future updates?
Thanks for your work on this project!