Creating a cisTarget database for Zebrafish

[RcuIT]

Hi
I managed to run create_cistarget_motif_databases.py for zebrafish and it generated three . feather files, zf.genes_vs_motifs.rankings.feather, zf.genes_vs_motifs.scores.feather and zf.motifs_vs_genes.scores.feather. How can I run create_cross_species_motifs_rankings_db.py ? because create_cistarget_motif_databases.py output did not generate cisTarget motifs vs regions or genes rankings databases. how can I continue with this? is it a must to have cross_species databases to run GRN ? Appreciate any response in this regard. thank you.

[ghuls]

For pySCENIC you only will need zf.genes_vs_motifs.rankings.feather. Cross-species it not necessary nowadays as it was in the past as we have way more motifs than we had in the past.

[RcuIT]

For pySCENIC you only will need zf.genes_vs_motifs.rankings.feather. Cross-species it not necessary nowadays as it was in the past as we have way more motifs than we had in the past.

Thanks for the clarification. I have another question. What should I pass to "org" parameter in the initializeScenic (). It says it needs, 'org' to be one of: mgi, hgnc, dmel. In the case of zebrafish, how could I specify the org?

[ghuls]

SCENIC is deprecated, use pySCENIC instead: https://pyscenic.readthedocs.io/en/latest/installation.html#docker-podman-and-singularity-apptainer-images pySCENIC does not need org as you just give a motif to TF annotation file with zebrafish genes as argument in the ctx step.

[RcuIT]

Thanks for the reply. I was trying to find a way to generate a motif to TF annotation file with zebrafish genes. The I found this database created by @stanaka6. I ran the following script, but got an empty .csv file.
singularity run -B /home/roshanpe/Seurat/data:/data aertslab-pyscenic-0.12.1.sif pyscenic ctx /home/roshanpe/Seurat/data/24_expr_mat.adjacencies.tsv /home/roshanpe/Seurat/data/zf.genes_vs_motifs.rankings.feather --annotations_fname /home/roshanpe/Seurat/data/Zebrafish_Motif2TF_db_ready_for_SCENIC.tbl --expression_mtx_fname /home/roshanpe/Seurat/data/24.exp.tsv --mode "custom_multiprocessing" --output /home/roshanpe/Seurat/data/regulons.csv --num_workers 6

while running the script, it shows the following for several genes.
2023-07-24 12:48:01,796 - pyscenic.transform - WARNING - Less than 80% of the genes in Regulon for si:dkey-253d23.9 could be mapped to zf.genes_vs_motifs.rankings. Skipping this module.

any suggestions to overcome this issue? Thanks

[ghuls]

while running the script, it shows the following for several genes.
2023-07-24 12:48:01,796 - pyscenic.transform - WARNING - Less than 80% of the genes in Regulon for si:dkey-253d23.9 could be mapped to zf.genes_vs_motifs.rankings. Skipping this module.
any suggestions to overcome this issue? Thanks

This means that your gene names in the database don't match the gene names in your expression matrix.

[RcuIT]

Thanks for the reply. Could you please provide me with steps to make a custom motif2TFs for zebrafish. I checked *resources.aertslab.org_cistarget_motif2tf_motifs-v9-nr.mgi-m0.001-o0.0.tbl.txt *file. I saw in the issue page you suggested replacing rat orthologs for mouse genes. I did try to do it for zebrafish. I downloaded zebrafish orthologs for mouse genes using ensembl. The above mouse txt file has around 163339 listed under the *gene_name* column. There I can find repeating of the same gene many times. Also some zebrafish genes are not having mouse ortholog and vice versa. How I can now replace zebrafish orthologs for mouse genes in the *gene_name* col. any special procedure to do this? The excel file here <https://upenn.app.box.com/file/1267957948488?s=ky3gu1ur4bl02n4c4jez3n33o8r8olvy> contains *zebrafish genes, mouse gene and genes* taking from gene_name col from the above tbl file. I would appreciate it if you could provide me with some info to replace zebrafish ortholog for mouse genes. Best, Roshan *________________________________________* *Roshan Priyarangana Perera, Ph.D (he/him/his)* Postdoctoral Research Scientist Department of Cell and Developmental Biology Biomedical Research Building II/III University of Pennsylvania Philadelphia, PA 19104, USA *Primary Contact* *Email :* ***@***.*** ***@***.***>* *Research lab :* *https://www.rajlab-neuro.com/ <https://www.rajlab-neuro.com/>* *Social Media* *linkedin : * *https://www.linkedin.com/in/roshan-priyarangana-perera-448ba340/* <https://www.linkedin.com/in/roshan-priyarangana-perera-448ba340/> *ResearchGate :* *https://www.researchgate.net/profile/Roshan-Perera-3 <https://www.researchgate.net/profile/Roshan-Perera-3>*

…

On Wed, 26 Jul 2023 at 11:05, Gert Hulselmans ***@***.***> wrote: while running the script, it shows the following for several genes. 2023-07-24 12:48:01,796 - pyscenic.transform - WARNING - Less than 80% of the genes in Regulon for si:dkey-253d23.9 could be mapped to zf.genes_vs_motifs.rankings. Skipping this module. any suggestions to overcome this issue? Thanks This means that your gene names in the database don't match the gene names in your expression matrix. — Reply to this email directly, view it on GitHub <#39 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AH4WDBZ5GMOFS7TRUIONSKDXSEW2VANCNFSM6AAAAAA2GHQBIY> . You are receiving this because you authored the thread.Message ID: ***@***.***>