"Anchors File Generated in Ortholog Mode Only Contains Three Columns, Preventing Downstream Visualization"

[SoledadPianist]

I'm encountering a problem when running the ortholog analysis with JCVI. I executed the following command:
python -m jcvi.compara.catalog ortholog Np-x sb --no_strip_names
However, the resulting anchors file only contains three columns (SpeciesA_gene, SpeciesB_gene, and score). Since the JCVI karyotype visualization module expects an anchors file with six columns (e.g., including start/end positions and orientation for each syntenic block),
the visualization step fails with an error like:
ValueError: not enough values to unpack (expected 6, got 3)

It seems that the ortholog command is outputting a simplified format by default, which does not include the full coordinate and orientation details required for downstream visualization. I have tried looking for options (like --simple=False) but haven't found a way to force a full-format (6-column) anchors file output from the catalog module.
Questions:

• Is there an available command-line parameter or workflow within JCVI that forces the generation of a full-format (6-column) anchors file when running the ortholog analysis?
• If not, what is the recommended procedure to obtain a compatible anchors file (with complete synteny block information) for use with JCVI karyotype?

Any help or guidance on resolving this issue for downstream visualization would be greatly appreciated.

[tanghaibao]

@SoledadPianist

The .anchors file is different from .simple file (which can be generated from the .anchors file.

Please follow the instructions here:
https://github.com/tanghaibao/jcvi/wiki/MCscan-(Python-version)#macrosynteny-visualization

[SoledadPianist]

@SoledadPianist

The .anchors file is different from .simple file (which can be generated from the .anchors file.

Please follow the instructions here: https://github.com/tanghaibao/jcvi/wiki/MCscan-(Python-version)#macrosynteny-visualization

I got what you mean.However,both the .anchorsand the.simple were only 3 column resulting in the visualization step error ValueError: not enough values to unpack (expected 6, got 3).Of course I had followed the instructions.That's why I wander if my input files have any problem or my command had somethting wrong.

[tanghaibao]

The .simple file should have 6 columns if it's generated correctly. You can run:

python -m jcvi.compara.synteny screen --minspan=30 --simple Np-x.sb.anchors Np-x.sb.anchors.new

This generates Np-x.sb.anchors.simple that you can use in the karyotype visualization.

[SoledadPianist]

The .simple file should have 6 columns if it's generated correctly. You can run:

python -m jcvi.compara.synteny screen --minspan=30 --simple Np-x.sb.anchors Np-x.sb.anchors.new

This generates Np-x.sb.anchors.simple that you can use in the karyotype visualization.

Thank you for help.You tips make sense.I’ve been following the JCVI tutorial for synteny visualization, and I noticed that the examples only demonstrate the use of the “g*” and “r*” prefixes (which map to green and red, respectively) for highlighting syntenic blocks. However, I encountered an issue when I attempted to use other prefixes, as my anchors file annotations with additional markers do not seem to be correctly mapped to valid colors(for example,I tried "b*" and "o*" for blue and orange which have get errors).

• Are there additional color mappings available in the current version of JCVI (for example, mappings for prefixes like “b*”, “y*”, etc.), or is the system limited to “g*” and “r*” only?
• If additional mappings are supported, is there a reference list of valid prefixes and their corresponding colors?
• If not, would you recommend a way to extend or customize the color mapping (perhaps by modifying the underlying source code) so that I can use other colors in my visualization?

[tanghaibao]

'o' is not supported. Here is all the colors allowed:

https://matplotlib.org/stable/gallery/color/named_colors.html

Other than named colors, you can also use hex color like #ff0000 (red).

[SoledadPianist]

'o' is not supported. Here is all the colors allowed:

https://matplotlib.org/stable/gallery/color/named_colors.html

Other than named colors, you can also use hex color like #ff0000 (red).

thank you

[SoledadPianist]

'o' is not supported. Here is all the colors allowed:

https://matplotlib.org/stable/gallery/color/named_colors.html

Other than named colors, you can also use hex color like #ff0000 (red).

I noticed that some of my target gene pairs are present in the full anchors file but get filtered out in the simple file (due to the minspan filter). I’d prefer not to lower minspan since that keeps out low-confidence blocks, but I really need those specific pairs to appear in the simple file for visualization. Manual editing is tough given the format differences. Is there any option or built-in feature to automatically add these target pairs to the simple file output？

[tanghaibao]

Unfortunately, there is no automated method, but you can try run the same command with and without --minspan, show the diff between the two .simple file, and then identify the pair that you want to visualize.

[SoledadPianist]

Unfortunately, there is no automated method, but you can try run the same command with and without --minspan, show the diff between the two .simple file, and then identify the pair that you want to visualize.

I have tried your tips.However,still many genes I hope to hilight did not exist in .anchor,.anchor.newand.anchor.simple.And some exist in .anchor.new but not in .anchorand.simple.
I wander what cause these problem.Also,the genes that I need to higlight were extracted from same cdsfile.