Dog genes

[bschilder]

Hey @igordot , hope all is well with you!

My colleagues and I noticed something odd in the Canis lupus familiaris orthologs in orthogene, and I traced it back to babelgene. See here for the full details: neurogenomics/orthogene#30

In the "symbol" column, I expect this to be the gene symbol as in the format of the origin species (in this case dog). This seems to be the case for other species like mouse, but not for dog.

source_id <- 9615
orths <- babelgene:::orthologs_df
orths <- subset(orths, taxon_id == source_id)
head(orths, 20)

Delving a bit deeper, i checked the number of human gene symbols that are identical to their non-human orthologs in each species. This varies quite a bit across species (even within mammals), but I was wondering if this is expected? I'm familiar with the different casing with mouse genes, but not sure what the convention is for the other species.

counts <- dplyr::group_by(babelgene:::orthologs_df, taxon_id) |>
    dplyr::summarise(identical_symbols=sum(human_symbol==symbol, na.rm = TRUE),
                     percent=sum(human_symbol==symbol, na.rm = TRUE)/dplyr::n()*100) |>
    dplyr::arrange(dplyr::desc(percent))
counts$species_name <- orthogene::map_species(counts$taxon_id, method = "babelgene")
counts

Repeated using the agr_orthologs_df data:

   counts_agr <- dplyr::group_by(babelgene:::agr_orthologs_df, species_name) |>
        dplyr::summarise(identical_symbols=sum(human_symbol==species_symbol),
                         percent=sum(human_symbol==species_symbol)/dplyr::n()*100) |>
        dplyr::arrange(dplyr::desc(percent))
counts_agr

Thanks!,
Brian

Session info

Using babelgene v22.9

Details

``` R version 4.2.1 (2022-06-23) Platform: x86_64-apple-darwin17.0 (64-bit) Running under: macOS Ventura 13.2.1

Matrix products: default
LAPACK: /Library/Frameworks/R.framework/Versions/4.2/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats graphics grDevices utils datasets methods base

other attached packages:
[1] ggtree_3.6.2 orthogene_1.5.1

loaded via a namespace (and not attached):
[1] utf8_1.2.3 devoptera_0.99.0 R.utils_2.12.2
[4] RUnit_0.4.32 tidyselect_1.2.0 RSQLite_2.3.0
[7] AnnotationDbi_1.60.2 htmlwidgets_1.6.2 grid_4.2.1
[10] combinat_0.0-8 devtools_2.4.5 munsell_0.5.0
[13] codetools_0.2-19 DT_0.27 miniUI_0.1.1.1
[16] withr_2.5.0 colorspace_2.1-0 Biobase_2.58.0
[19] filelock_1.0.2 TreeTools_1.9.1 knitr_1.42
[22] rstudioapi_0.14 stats4_4.2.1 ggsignif_0.6.4
[25] MatrixGenerics_1.10.0 Rdpack_2.4 labeling_0.4.2
[28] GenomeInfoDbData_1.2.9 mnormt_2.1.1 optimParallel_1.0-2
[31] topGO_2.50.0 bit64_4.0.5 farver_2.1.1
[34] rprojroot_2.0.3 coda_0.19-4 vctrs_0.6.1
[37] treeio_1.23.1 generics_0.1.3 clusterGeneration_1.3.7
[40] xfun_0.37 BiocFileCache_2.6.1 R6_2.5.1
[43] doParallel_1.0.17 GenomeInfoDb_1.34.9 rsvg_2.4.0
[46] grImport2_0.2-0 bitops_1.0-7 cachem_1.0.7
[49] gridGraphics_0.5-1 DelayedArray_0.24.0 promises_1.2.0.1
[52] scales_1.2.1 gtable_0.3.3 biocViews_1.66.3
[55] processx_3.8.0 phangorn_2.11.1 ontologyPlot_1.6
[58] rlang_1.1.0 scatterplot3d_0.3-43 rstatix_0.7.2
[61] lazyeval_0.2.2 broom_1.0.4 BiocManager_1.30.20
[64] yaml_2.3.7 abind_1.4-5 ggimage_0.3.1.002
[67] backports_1.4.1 httpuv_1.6.9 RBGL_1.74.0
[70] tools_4.2.1 usethis_2.1.6 ggplotify_0.1.0
[73] ggplot2_3.4.1 ellipsis_0.3.2 paintmap_1.0
[76] RColorBrewer_1.1-3 BiocGenerics_0.44.0 sessioninfo_1.2.2
[79] Rcpp_1.0.10 plyr_1.8.8 base64enc_0.1-3
[82] zlibbioc_1.44.0 purrr_1.0.1 RCurl_1.98-1.10
[85] ps_1.7.3 prettyunits_1.1.1 ggpubr_0.6.0
[88] urlchecker_1.0.1 S4Vectors_0.36.2 SummarizedExperiment_1.28.0
[91] grr_0.9.5 fs_1.6.1 here_1.0.1
[94] magrittr_2.0.3 data.table_1.14.8 magick_2.7.4
[97] SparseM_1.81 R.cache_0.16.0 matrixStats_0.63.0
[100] pkgload_1.3.2 patchwork_1.1.2 mime_0.12
[103] evaluate_0.20 xtable_1.8-4 XML_3.99-0.14
[106] jpeg_0.1-10 IRanges_2.32.0 compiler_4.2.1
[109] tibble_3.2.1 maps_3.4.1 crayon_1.5.2
[112] R.oo_1.25.0 htmltools_0.5.4 ggfun_0.0.9
[115] later_1.3.0 tidyr_1.3.0 aplot_0.1.10
[118] expm_0.999-7 ontoProc_1.20.0 DBI_1.1.3
[121] gprofiler2_0.2.1 dbplyr_2.3.2 MASS_7.3-58.3
[124] rappdirs_0.3.3 babelgene_22.9 Matrix_1.5-3
[127] car_3.1-1 piggyback_0.1.4 cli_3.6.1
[130] quadprog_1.5-8 R.methodsS3_1.8.2 rbibutils_2.2.13
[133] parallel_4.2.1 igraph_1.4.1 GenomicRanges_1.50.2
[136] pkgconfig_2.0.3 numDeriv_2016.8-1.1 plotly_4.10.1
[139] foreach_1.5.2 stringdist_0.9.10 XVector_0.38.0
[142] BiocCheck_1.34.3 yulab.utils_0.0.6 stringr_1.5.0
[145] callr_3.7.3 digest_0.6.31 phytools_1.5-1
[148] graph_1.76.0 Biostrings_2.66.0 rmarkdown_2.20.1
[151] fastmatch_1.1-3 tidytree_0.4.2 curl_5.0.0
[154] shiny_1.7.4 lifecycle_1.0.3 nlme_3.1-162
[157] jsonlite_1.8.4 carData_3.0-5 OmaDB_2.14.0
[160] viridisLite_0.4.1 fansi_1.0.4 pillar_1.9.0
[163] ontologyIndex_2.10 lattice_0.20-45 homologene_1.4.68.19.3.27
[166] KEGGREST_1.38.0 fastmap_1.1.1 httr_1.4.5
[169] plotrix_3.8-2 pkgbuild_1.4.0 GO.db_3.16.0
[172] interactiveDisplayBase_1.36.0 glue_1.6.2 remotes_2.4.2
[175] png_0.1-8 iterators_1.0.14 BiocVersion_3.16.0
[178] bit_4.0.5 Rgraphviz_2.42.0 stringi_1.7.12
[181] profvis_0.3.7 blob_1.2.4 AnnotationHub_3.6.0
[184] rphylopic_1.0.0 memoise_2.0.1 dplyr_1.1.1
[187] ape_5.7-1

</details>

[bschilder]

Ok, so it turns out the reason for my original issue was actually separate from the number of orths with identical names.

Instead, it had to do with the number of 1:1 orthologs as counted by symbol:symbol as opposed to symbol:ensemblID
neurogenomics/orthogene#30 (comment)

[igordot]

So should I consider this issue resolved?

It sounds like the problem was with many human Ensembl IDs with two matches in the dog. I am not sure how that occurred.

[bschilder]

So should I consider this issue resolved?

It sounds like the problem was with many human Ensembl IDs with two matches in the dog. I am not sure how that occurred.

In my particular use case it does resolve the issue. But I think it would still be good to address the following questions:

1. Are the genes that have identical symbols across species is simply a reflection of how those non-human genomes are annotate, and not an artifact accidentally introduced when preparing the files.
2. Why are there less 1:1 orthologs when considering Ensembl IDs instead of gene symbols? Might not be something we can answer right away, but would be good to know for both our packages! I can help look into this as well.
3. For the babelgene::orthologs functions, would it be worth adding a note to the documentation regarding the (potentially big) difference between Ensembl ID-based 1:1 orthologs and gene symbol-based 1:1 orthologs? Might not be something people are generally aware of when using the output in downstream analyses.