is it possible to plot a list of GO terms?

[ghost]

Hi,

I want to visualize the structural (spatial) relationship for a list of GO terms in the GO hierarchy (for example, MF), is it possible to use your code for that purpose? e.g. python plot_go_term.py --term_list_file term_list.txt?

Thanks!

[dvklopfenstein]

Thank you for your interest in GOOATOOLS and taking the time to write us expressing interest in an important feature.

Yes! We are adding the ability to read a list of GO IDs from a file soon in a GO plotting script which will take advantage of new (hopefully) soon-to-published functionality in GOATOOLS.

For example, to visually confirm that we correctly set up a test to read up and down the GO hierarchy using both the standard _isa (black arrows) and the optional relationship attributes(dashed magenta arrows), we do a plot using this text file containing a list of GO IDs and colors:

#-------------------------------------------
# Source for tests up GO hierachy
GO:0007389#fffdda # Yellow

# Ancestors up hierarchy through is_a attribute
GO:0008150#e0ffdc # Green
GO:0032501#e0ffdc # Green

# Ancestors up hierarchy through relationship attributes
GO:0007275#deddff # Purple 
GO:0048856#deddff # Purple 
GO:0032502#deddff # Purple

#-------------------------------------------
# Source for tests down GO hierarchy
GO:0003143#fffdda # Yellow  embryonic heart tube morphogenesis

# Ancestors up hierarchy through is_a attribute
GO:0003146#e0ffdc # Green

# Ancestors up hierarchy through relationship attributes
GO:0003304#deddff # Purple

And run this command:

$ go_plot -i test_gos.txt --obo=../goatools/tests/data/heartjogging.obo -r -o heartjogging_test.png

And see this plot:

Here is the help doc for the script using new functionality:

Command-line script to create GO term diagrams

Usage:
  go_plot.py [GO ...] [options]

Options:
  -h --help                            show this help message and exit
  -i --go_file=<file.txt>              GO IDs in an ASCII file
  -o <file.png>, --outfile=<file.png>  Plot file name [default: go_plot.png]
  -r --relationship                    Plot all relationships
  -s <sections.txt> --sections=<sections.txt>  Sections file for grouping
  -S <sections module str>             Sections file for grouping

  --gaf=<file.gaf>                     Annotations from a gaf file
  --gene2go=<gene2go>                  Annotations from a gene2go file downloaded from NCBI

  --obo=<file.obo>                     Ontologies in obo file [default: go-basic.obo].

  -t <title>, --title=<title>          Title string to place in image
  -p --parentcnt                       Include parent count in each GO term
  -c --childcnt                        Include child count in each GO term
  --shorten                            Shorten the GO name on plots
  --mark_alt_id                        Add 'a' if GO ID is an alternate ID: GO:0007582a
  --draw-children                      Draw children. By default, they are not drawn.
  --go_aliases=<go_aliases.txt>        ASCII file containing letter alias

  --norel                              Don't load relationship from the GO DAG

This code will be released soon.

Thank you very much for your interest in GOATOOLS.

[dvklopfenstein]

FYI, To compute the Resnik and Lin similarities between pairs of these GOs:

human Information content 5.682202 GO:0015318 inorganic molecular entity transmembrane transporter activity
human Information content 4.660596 GO:0140096 catalytic activity, acting on a protein
human Information content 7.125990 GO:0140097 catalytic activity, acting on DNA
human Information content 6.479823 GO:0140098 catalytic activity, acting on RNA
human Information content 7.622427 GO:0140101 catalytic activity, acting on a tRNA
human Information content 4.787814 GO:0140110 transcription regulator activity

Like this:

GO #1      GO #2      Resnik Lin
---------- ---------- ------ -------
GO:0015318 GO:0140096 2.6690 -0.5120
GO:0015318 GO:0140097 2.6690 -0.4628
GO:0015318 GO:0140098 2.6690 -0.4815
GO:0015318 GO:0140101 2.6690 -0.4369
GO:0015318 GO:0140110 2.6690 -0.4828
GO:0140096 GO:0140097 3.3710 -0.6298
GO:0140096 GO:0140098 3.3710 -0.6574
GO:0140096 GO:0140101 3.3710 -0.5921
GO:0140096 GO:0140110 2.6690 -0.5220
GO:0140097 GO:0140098 3.3710 -0.5933
GO:0140097 GO:0140101 3.3710 -0.5395
GO:0140097 GO:0140110 2.6690 -0.4709
GO:0140098 GO:0140101 5.4578 -0.9061
GO:0140098 GO:0140110 2.6690 -0.4903
GO:0140101 GO:0140110 2.6690 -0.4442

See new test: ./tests/test_semantic_similarity_best4lex.py

[dvklopfenstein]

I will close this issue when our publication is out and our unpublished code is released.

[ghost]

Okay, I am looking forward to the new functionality! Thank you!

--

Linhua (Alex) Wang

Bioinformatician

Department of Genomic Science/Genetics

Icahn School of Medicine at Mount Sinai, New York

On Sat, May 19, 2018 at 1:37 PM, DV Klopfenstein notifications@github.com wrote:

Thank you for your interest in GOOATOOLS and taking the time to write us expressing interest in an important feature.

Yes! We are adding the ability to read a list of GO IDs from a file soon in a GO plotting script which will take advantage of new (hopefully) soon-to-published functionality in GOATOOLS.

For example, to visually confirm that we correctly set up a test to read up and down the GO hierarchy using both the standard is_a and the optional relationship attributes, we do a plot using this text file containing a list of GO IDs and colors:

-------------------------------------------

Source for tests up GO hierachy

GO:0007389#fffdda # Yellow

Ancestors up hierarchy through is_a attribute

GO:0008150#e0ffdc # Green GO:0032501#e0ffdc # Green

Ancestors up hierarchy through relationship attributes

GO:0007275#deddff # Purple GO:0048856#deddff # Purple GO:0032502#deddff # Purple

-------------------------------------------

Source for tests down GO hierarchy

GO:0003143#fffdda # Yellow embryonic heart tube morphogenesis

Ancestors up hierarchy through is_a attribute

GO:0003146#e0ffdc # Green

Ancestors up hierarchy through relationship attributes

GO:0003304#deddff # Purple

And run this command:

$ go_plot -i test_gos.txt --obo=../goatools/tests/data/heartjogging.obo -r -o heartjogging_test.png

And see this plot: [image: heartjogging_test] https://user-images.githubusercontent.com/7278188/40271267-e639fdac-5b68-11e8-9727-4de136540787.png

Here is the help doc for the script using new functionality:

Command-line script to create GO term diagrams

Usage: go_plot.py [GO ...] [options]

Options: -h --help show this help message and exit -i --go_file= GO IDs in an ASCII file -o , --outfile= Plot file name [default: go_plot.png] -r --relationship Plot all relationships -s --sections= Sections file for grouping -S Sections file for grouping

--gaf= Annotations from a gaf file --gene2go= Annotations from a gene2go file downloaded from NCBI

--obo= Ontologies in obo file [default: go-basic.obo].

-t