Some questions about the CellNEST

[litun-fkby]

Hi ,
Thanks for providing this tool , and when i use the CellNEST i meet some problem， so i have some questions.

I ues the publicly available Visium sample on human lymph node from https://cf.10xgenomics.com/samples/spatial-exp/1.1.0/V1_Human_Lymph_Node/V1_Human_Lymph_Node_filtered_feature_bc_matrix.h5 and https://cf.10xgenomics.com/samples/spatial-exp/1.1.0/V1_Human_Lymph_Node/V1_Human_Lymph_Node_spatial.tar.gz as input , and use the Singularity Image to run , the command is:

SCRIPT_PATH=$(which cellnest)
SCRIPT_DIR=$(dirname $SCRIPT_PATH)
log_dir='log.info'
if [ ! -d $log_dir ]
then
mkdir -p $log_dir
fi

preprocess

cellnest preprocess --data_name='V1_Human_Lymph_Node_spatial' --data_from='data/human_lymph_node' --database_path="$SCRIPT_DIR/database/CellNEST_database.csv" --split=1 --filter_min_cell=1 > $log_dir/preprocess.log

list1="1"
epoch_number=8000
for number in ${list1[@]}
do
cellnest run --data_name='V1_Human_Lymph_Node_spatial' --num_epoch $epoch_number --manual_seed='yes' --seed=$number --model_name='CellNEST_V1_Human_Lymph_Node_spatial' --run_id=$number --total_subgraphs=3 > $log_dir/train_model_run$number.log
done

postprocess

cellnest postprocess_split --data_name='V1_Human_Lymph_Node_spatial' --model_name="CellNEST_V1_Human_Lymph_Node_spatial" --total_runs=1 --total_subgraphs=3 > $log_dir/postprocess.log

visualize

cellnest visualize --data_name='V1_Human_Lymph_Node_spatial' --model_name='CellNEST_V1_Human_Lymph_Node_spatial' --top_edge_count=40000 --annotation_file_path='data/human_lymph_node/V1_Human_Lymph_Node_spatial_annotation.csv' > $log_dir/Visualize.log

the visualize output component_plot.html is different from the github, my result has 4 component but 17 component in github，and the CCC in histogram_byFrequency_plot.html also is different:
my result:

the github result:

question 1: What suggestions do you have here for the result?
question 2: how can i get the plot or mygraph.html result like fig.1h and fig.5h in paper for the clear and attractive result? The mygraph.html use the "cellnest visualize" will be like this:

Some points and lines overlap each other. When magnified, the communication signal also overlaps with the spot points. It's better for visualize improve and add the component lab or other lab.

The "cellnest relay_extract" command , when i ues the option --target_relay="CLSTN1-CD47 to CCL21-CXCR4" to get the target relay visualize result , and get the error , because the parameters have spaces ,the script in extract_relay_cellnest.py maybe change like this:

and the question3: is there any script for relay network visualize like "cellnest visualize "? The script can filter result by component , cell type and relay network.

Hope for your reply!
Thanks!