geneontology / helpdesk

The Gene Ontology Helpdesk

Helpdesk

• Resources GO Home FAQs Questions/issues/suggestions Status

The GO Helpdesk

Welcome to the Gene Ontology Helpdesk.

Do you have general questions about GO? Check our Frequently Asked Questions, as your query may have already been answered.

Do you have a specific question about the GO, its data, usage, software, or services? Create a ticket on our discussion tracker and join the GO community on GitHub!

Are you having trouble using or connecting to any of our public services? Check our status page.

Are you interested in contributing to annotations or the ontology? Please visit the guide on Contributing to GO.

Join our mailing lists and stay informed! The GOC supports several mailing lists to foster discussion within the GO community. Subscribe to our public GO-friends list to follow general developments, discussions about the GO, or to post a message to the GO community. Contributing members of the GOC should contact us to ensure they have been added to the correct lists.

• GO-friends mailing list

Still have a question or comment? Please send us an email and we will respond as soon as possible.

Best wishes, -The GO Helpdesk Team

Copyright © 1999-2024 the Gene Ontology (CC-BY 4.0) The Gene Ontology Consortium is supported by a P41 grant from the National Human Genome Research Institute (NHGRI) [grant U41 HG002273].

helpdesk's Issues

GO Annotation files from FlyBase etc have different number of annotated genes than GO Consortium - Which one to choose

As the title notes, an example is that FlyBase contains ~1000 lethal knockout genes which are annotated with a lethal phenotype and a GO term. Whereas if I use the go_annotation file from GO Consortium, there is only ~500 which are annotated with a GO term.
Is FlyBase more liberal or is the annotation file from the Go Consortium older?

Mouse annotation file

Dear GO Team,

Could you please let me know what version of Mouse genome the current annotation (24426 records) is corresponding to?

Thanks.

Kind regards,

C.

Bug in go.obo?

Hi, I think I found a small bug in the file http://www.geneontology.org/ontology/go.obo. Could you confirm or explain to me why this is intended behaviour?

The change is GOC: bmeldal to GOC:bmeldal without the space. I think in OBO v1.2, after the space in a dbxref there has to be a quoted string. I think here, the space is incorrect.

Cheers!

--- go.obo	2017-06-16 15:33:39.738711044 +0200
+++ go.obo	2017-06-16 15:33:43.406695176 +0200
@@ -479648,7 +479648,7 @@
 id: GO:0106003
 name: amyloid-beta complex
 namespace: cellular_component
-def: "Protein complex involved in modulation of signaling and synaptic function in the brain, predominantly in the cerebral cortex and hippocampus. Forms dimers and multimers of amyloid beta peptide 40 and peptide 42 (proteolytic cleavage products of amyloid beta A4 protein, also known as amyloid beta precursor protein). Mostly found in the extracellular space with a proportion occurring as membrane-bound species. Influences synaptic plasticity through various receptors, mediates dendritic spine loss leading to decreased synapse density, inhibits long-term potentiation (LTP) and enhances long-term depression (LTD). Soluble multimeric form is the main pathogenic species linked to Alzheimer's disease." [GOC: bmeldal, PMID:18568035]
+def: "Protein complex involved in modulation of signaling and synaptic function in the brain, predominantly in the cerebral cortex and hippocampus. Forms dimers and multimers of amyloid beta peptide 40 and peptide 42 (proteolytic cleavage products of amyloid beta A4 protein, also known as amyloid beta precursor protein). Mostly found in the extracellular space with a proportion occurring as membrane-bound species. Influences synaptic plasticity through various receptors, mediates dendritic spine loss leading to decreased synapse density, inhibits long-term potentiation (LTP) and enhances long-term depression (LTD). Soluble multimeric form is the main pathogenic species linked to Alzheimer's disease." [GOC:bmeldal, PMID:18568035]
 comment: An example is Protein 40 of APP (P05067-PRO_0000000093) in PMID:18568035 (inferred by direct assay).
 synonym: "Abeta complex" EXACT []
 synonym: "Abeta-derived diffusible ligand complex" NARROW []

Which Gene ontology database tables contains NCBI sequence accession numbers or Entrez id

I am making a pipeline for GO analysis (starting from sequence). I dont want to use blast2GO. I have Gene ontology local database and i want to map blast output to gene ontology categories but i am not able to find which Gene ontology database tables contains NCBI sequence accession numbers or Entrez id so that to map the blast output to GO categories

secretome and membrane receptor profiles

Can we use GO to give a satisfactory answer for:

https://bioinformatics.stackexchange.com/questions/2241/secretome-and-membrane-receptor-profiles/2268

Why Epidermal differentiation complex is not available in GO?

I see epidermal differentiation complex which has been studied quite extensively. Is there a reason why it is not included in GO? Thanks.

https://en.wikipedia.org/wiki/Epidermal_differentiation_complex

Improve information on citing the GO in analytical publications

Provide easy to locate (obvious) information describing what details need to be included in any publications that utilized GO in their analysis. This includes the GOC paper to reference, datas, use of NOTs, etc.

TermGenie

Hi,
Is the TermGenie site still down? Is there an alternative?
Thanks
Nancy

GO annotation don't work

Hello!

GO annotation link http://amigo.geneontology.org/amigo/landing don't work. :(

Creating a user defined GOSlim

Hello,
I have some proteomics data in the context of wound healing and inflammation. To categorise these proteins, I would like to create a user-defined GOslim for this specific context.

I am quite new to GO and am unsure of what the current tools available for creating GOslims are. Is there a method for creating a GOslim without too much user bias? For example, taking the most enriched categories in the dataset and using those as GOslim terms?

My goal is to create a histogram with the number of proteins up- and down-regulated in each GOslim category. I just need to create those categories first.

Advice would be much appreciated!

accessing all annotated GO terms IDs (Including ancestors) for a gene programatically

One of our users has asked how to do this as part of some protein-protein interaction prediction method.

I have seen it answered previously on the GO helpdesk more than once but I don't see it in the FAQ . Could it be added?

Thanks,

VAl

Where is the antimicrobial peptide GO term?

S100A7 is an antimicrobial peptide. But I don't find such an annotation in GO. Is it available?

Pfam2go download is corrupt

The pfam2go file http://geneontology.org/external2go/pfam2go is just an error message.
interpro2go, smart2go, and other files updated on Jul 16 have the same issue.

Can it be fixed? Thanks.

Access denied on a public URL

Click on http://www.geneontology.org/page/tools (which is a link on the site)
result is

Access denied
You are not authorized to access this page.

Expected result: a nice web page describing all the wonderful third party tools or a message saying that this is now discontinued and the user should search at ....

Ted

number of curated papers

Hi, I'm trying to estimate how many papers have been curated for GO. This can be broken down into several more specific questions:

1. how many papers contain EXP annotations.
   http://amigo.geneontology.org/amigo/base_statistics#experimental-annotation-publications-by-assigner

The last figure on this page has some numbers, but they appear to be high. For example, according to Fig 3, there are around 700k experimental annotations. If we add up the # of papers in the last figure, it's around 1million, less than the # of total EXP annotations.

2. is there any estimate on how many papers contain gene function information?

I'm trying to scope out how much work remaining to be done for GO annotation. Thanks,

Donghui

Question about a GOOSE query

Hi,
I am interested in using GOOSE to get a list of all human gene products annotated to a selected GO term and all of its decendant terms. I'd also like to be able to specify the evidence codes that I would want to include:
(IPI) (IDA) (EXP) (IC) (TAS) (IMP)(IGI). Would you be able to help me design such a query?
Best regards,
Lisa

newer COG2go version?

Hi there,
I'm looking for a newer version of COG2go database, do you know where I can get them?
Thanks,
arolnij

Can I perform term enrichment analysis for a species not listed in the enrichment tool?

Hi
I am working on Streptococcus agalactiae and streptococcus equi, is there any chance for me to perform Enrichment analysis because I couldn't find the organisms in the list.
thanks

GO accession IDs for GO Molecular Function gene sets

I'm trying to find GO accession IDs for all the gene sets in the GO Molecular Function category. It seems like a very straightforward query, but I've spent several hours on it trying GO ftp service, Amigo, Ensembl Biomart, all without success. Please suggest a solution.

About statistics of available lncRNA annotations in current GOA

Hi,

Currently we need some statistics regarding the lncRNA annotations available through GO; specifically, we'd like to obtain a list of lncRNAs annotated on GO (from all possible species), along with their annotations.

I tried to search for some lncRNA (e.g., Xist, Hotair) and found some hits, but we don't know how to find all of them. I also tried to download the entire GOA dataset and find out whether I could extract all lncRNA records; while I saw a number of lncRNAs have in their full names "non protein-coding" there are some lncRNAs that cannot be detected as lncRNAs in this way (e.g. the Ftx transcript from Rat.)

Could you please tell us how we could obtain such a list?

Thanks

Use map2slim without input GAF

As per email:

Hi Anna,

You should be able to run the go slim mapper Maxim pointed to at https://github.com/EBISPOT/slim-o-matic/tree/master/go_slim_mapper

A few notes:

• We were using a specific GO obo file (in the GH repo) as we wanted to slim only against some relations. IIRC we excluded regulates. As indicated at https://github.com/owlcollab/owltools/wiki/Map2Slim, you could also use the go basic version at http://purl.obolibrary.org/obo/go/go-basic.obo
• You need to have OWL tools installed to run map2slim. The version we used is also in the GH repo (so downloading the whole folder should just work)
• test_input.txt: a file which contains the GO IDs, one per line, that you want to slim (i.e., in your case what you call the "children")
• The script generates the file pipeline_input_annotations.gaf. As you can see at https://github.com/EBISPOT/slim-o-matic/blob/master/go_slim_mapper/pipeline_input_annotations.gaf this is a "fake" GAF file which contains associations to the same gene throughout (same PMID, evidence etc as well). The only difference between the lines are the GO IDs to map.
• metagenomics_go_slim_ids.txt contains the IDs you want to slim against (i.e., the "parents")
• the script will use the fake GAF and the list of slim IDs to generate the file pipeline_mapped_annotations.gaf as shown at https://github.com/EBISPOT/slim-o-matic/blob/master/go_slim_mapper/pipeline_mapped_annotations.gaf (first column contains the GO IDs of the children, 4th column contains the GO IDs they'd be mapped against)

I hope this helps, let me know if you have any issue (or if everything works well ;))

Kind regards,
Melanie

On 04/07/2017 20:09, Chris Mungall wrote:

Hi Anna

I'll take a look at this later, can you file a ticket here to make sure we don't forget about this, thanks!
https://github.com/geneontology/helpdesk/issues/new

On 4 Jul 2017, at 6:47, Anna Esteve Codina wrote:

Melanie,

It would be helpful to have the code to generate the fake association file (whenever possible!).

I knew Map2Slim but I discarded the option as I should provide gene ID/names.

Thanks!

 

Anna Esteve Codina, PhD

Asunto: Re: [go-discuss] children to GOslim without genes

 

Hi Anna,

 

When I wanted to do that in the past I created a "fake" gene association file with the GO IDs. I was then able to pass this and the GO slim IDs to the map2slim tool at https://github.com/owlcollab/owltools/wiki/Map2Slim

 

I am on leave and only have my phone with me today, but if needed I have some code to generate the "fake" file and can find it later this week when back at my computer; it was actually written by Maxim (cc'ed) who may also be able to help in the meantime.

 

Cheers,

Melanie




On Jul 4, 2017, at 9:46 AM, Anna Esteve Codina  wrote:

    Dear all,

     

    I am interested in mapping specific children GOs to GOslim but I do not have gene IDs.

    Is there a tool to do that?

    Thanks,

Wrong media type for .obo downloads

The media type for http://purl.obolibrary.org/obo/go.obo is rdf+xml, which is wrong. I don't know what it should be - text perhaps - but it shouldn't be rdf+xml.

bash-3.2$ curl -I -L http://purl.obolibrary.org/obo/go.obo
HTTP/1.1 302 Found
Date: Sat, 05 Aug 2017 12:03:48 GMT
Server: Apache/2.4.7 (Ubuntu)
Location: http://snapshot.geneontology.org/ontology/go.obo
Content-Type: text/html; charset=iso-8859-1

HTTP/1.1 200 OK
Content-Type: application/rdf+xml
Content-Length: 36388562
Connection: keep-alive
Date: Thu, 03 Aug 2017 17:04:18 GMT
Last-Modified: Thu, 03 Aug 2017 14:18:37 GMT
ETag: "cc9fa236353864d31cd2af755337253e-3"
x-amz-meta-s3cmd-attrs: uid:114/gname:skyhook/uname:jenkins/gid:1003/mode:33188/mtime:1501744075/atime:1501744083/md5:6829075a7f7b950c08509ad414113dfa/ctime:1501744075
Accept-Ranges: bytes
Server: AmazonS3
X-Cache: RefreshHit from cloudfront
Via: 1.1 72882d2d20025ce740b1efae5c3e8544.cloudfront.net (CloudFront)
X-Amz-Cf-Id: 40KAOVo0BLaF2pDhpMV0uQaA38gOErcmDsRioxvHX5hPHVtNT3ZAaw==

bash-3.2$ 

Some Questions About .obo Format Files

Dear Professor,
Hello!
Now, I research that “Comprehensive Identification of Cell Fate Decision-related ncRNAs in Human”. So far, I’ve read your paper “OBO-Edit—an ontology editor for biologists.”.
I'm very interested in your work that “OBO-Edit—an ontology editor for biologists.”. For identification of ncRNA (particularly lncRNA and miRNA), we need firstly that systematically select core reprogramming-related transcript factors (TFs). Then, identifying ncRNAs that associated with these TFs. To calculate differential expression genes, I need two classes of sample: the first is case, and the second is control. I want to class samples by cell type's .obo format files.
However, I don’t know how to operate cell type tree file (.obo format file, ff-phase1-prerelease-20130717.obo). The people around me also did not have the experience of dealing with thus data.
I think it's very helpful for my experiments with your help. Because it’s the beginning of our research. Do you have some recommended software packages or papers for dealing with .obo format files to calculate differential expression for me? It’s not possible to manually handle the cell types, because of the number of human cell types in our research is more than 800.
I really appreciate your positive response.
Thank you very much! Good Luck!
Best Regards.
Liuzhe

Size of go_monthly-assocdb-data.gz

Hi,
The size of go_monthly-assocdb-data.gz dated 07-Mar-2017 is 228M, whereas that of the same file dated 07-Jan-2017 is 6.2G.
Is this file correct?

Thanks.

PANTHER webiste down for maintenance

Hi,

Since Satruday the PANTHER website seems to be down for maintenance (http://pantherdb.org) and
I was wondering whether you were aware of this as when I check your status page it seems to indicate everything is working fine.
Do you have any predictions of when the service will be resumed?

Thank you in advance,

Best,

Ana

GO powerpoint presentations

Hello there,

A collection of old GO ppts were available on the wiki or the website. I can't locate them. Can you please point me to those?

Thanks,
Rama

GO's fully update

Hi,

I noticed http://archive.geneontology.org/latest-full/go_monthly-assocdb-tables.tar.gz is not been updated for several months now (the last update was in JAN). I am wondering if there will be updates soon.

Thanks,

Anney

Can filenames use the same pattern?

All (gene association) filenames have the same pattern for their name:

gene_association.[SPECIES|ORG].gz

Except for the ones from the EBI. Can they all use the same pattern instead? Or is there a specific reason for the difference? It would make my life a lot easier! :)

Thanks!

Refining a search

From the now-archived JIRA GO help desk.

From : Lovisa Åkesson

Hi,
I'm in the need of estimating the number of human genes localizing to multiple cellular compartments (or organelles) based on the gene ontology cellular compartment terms. However, when I download the terms for each gene, most of the GO-terms are redundant. Is there a way to group all the terms for one gene into one major organelle or cellular compartment? I have tried to group them to their parent term before, but most of the parent terms are not on the same hierarchy level for all organelles which makes it hard.
Thanks in advance!
Kind regards,
Lovisa

Change GO:0033309 Description

Hi, a user [who] wrote into SGD helpdesk about a change to this description (GO:0033309 SBF transcription complex). See below. Can someone please change this.

Thanks,
Sage Hellerstedt (SGD, Biocuration Assistant)

"Description: A protein complex that binds to the Swi4/6 cell cycle box (SCB) promoter element, consensus sequence CRCGAAA, and activates transcription during the G1/S transition of the cell cycle. In Saccharomyces, the complex contains a heterodimer of the DNA binding protein Swi6p and the activator Swi4p, and is associated with additional proteins known as Whi5p and Msa1p.

It should read:
Description: A protein complex that binds to the Swi4/6 cell cycle box (SCB) promoter element, consensus sequence CRCGAAA, and activates transcription during the G1/S transition of the cell cycle. In Saccharomyces, the complex contains a heterodimer of the DNA binding protein Swi4p and the activator Swi6p, and is associated with additional regulatory proteins known as Whi5p and Msa1p."

latest-full/go_monthly-assocdb-tables.tar.gz

The following file is dated Jan 2017:

http://archive.geneontology.org/latest-full/go_monthly-assocdb-tables.tar.gz

In prior years, this file was updated 5-6 times per year. Has the policy changed, or is there some technical problem with monthly updates?

Thanks !

Interpro references disappeared in lite/go_weekly-assocdb.rdf-xml

It seems that all but 1 Interpro references have disappeared from the lite/go_weekly-assocdb.rdf-xml
data file in the latest release:
ftp://ftp.geneontology.org/go/godatabase/archive/lite/2017-04-29/go_weekly-assocdb.rdf-xml.gz
They are present in the data file from the week before:
ftp://ftp.geneontology.org/go/godatabase/archive/lite/2017-04-22/go_weekly-assocdb.rdf-xml.gz

Similarly all references to Prints, Prosite, Prodom, Pfam, Smart and Pirsf are also missing in the latest datafile.

Is this by design, or did something go wrong in the data build ?

GO enrichment

Hi,

I have a list of GO1 (associated to amino acid Sequences of interest) and a list of GO2 (associated to Subjects aligned to the sequences of interest by blastp).

There's no specific organism (either Eukaryota, either Bacteria)

Would there be a tool in order to compare the GO enrichment between the two GO sets or better between each GO set and the background ? In order to see if one set is more enriched in regulatory networks, metabolism, ...

Thank you
Kind regards
PS: The equivalent would be a topGO analysis but for sure TopGO analysis is not possible to apply in our case. I don't start from gene lists. I start with GO lists (GO associated to amino-acid sequences ).

Error in GO 0097710 viral terminase, small subunit

The PT for this concept specifies the small subunit, but it also has an SY that is the large subunit. Perhaps this is a typo and should read "small" for the SY as well:

viral terminase, small subunit [GO2017_04_28/PT/GO:0097710]
virus terminase, large subunit [GO2017_04_28/SY/GO:0097710]

There is another existing concept for the large subunit:

viral terminase, large subunit [GO2017_04_28/PT/GO:0098009]
virus terminase, large subunit [GO2017_04_28/SY/GO:0098009]

Thanks,
Miranda Jarnot
UMLS, NLM

GO ftp latest-full MySQL dump not available?

Dear GO-Team,

I was wondering what happened to the files in:
http://archive.geneontology.org/full/latest/
Currently March 2017.

The file assocdb-data only weights about 230 MB compared to 6.2 GB in January.
From the table sizes I get the impression that the "association" table is incomplete (March 2017).

The February folder is completely empty.

Am I missing something, or has the file's content changed?

I'd appreciate it if you could explain what happened here.

Best,
David

How to find all hormone receptors?

Are you aware of a complete list of hormone receptors?

It seems that GO does not have a term for all hormone receptors. (If so, should GO add such a GO term?)

The wiki page is incomplete.

https://en.wikipedia.org/wiki/Hormone_receptor

The following is a database of hormone receptors (the symbols are pasted below). But I am not sure how complete it is.

http://crdd.osdd.net/raghava/hmrbase/nphormone.php?table=recep&db=hmrbase&f_name=origin_org&f_value=Homo%20sapiens%20(Human)

Do you have any suggestions on where to find the complete list of
hormone receptors?

ADGRE1
ADGRE2
ADGRE3
ADRA1A
ADRA1B
ADRA1D
ADRA2A
ADRA2B
ADRA2C
ADRB1
ADRB2
ADRB3
AMHR2
AR
ATP6AP2
AVPR1A
AVPR1B
AVPR2
CALCR
CALCRL
CCKAR
CCKBR
CRHR1
CRHR2
CYSLTR1
CYSLTR2
ESR1
ESR2
ESRRA
ESRRB
ESRRG
FPR2
FSHR
GCGR
GHR
GHRHR
GHSR
GLP1R
GLP2R
GNRHR
GPR182
GPR50
GRPR
HNF4A
HNF4G
HTR1A
HTR1B
HTR1D
HTR1E
HTR1F
HTR2A
HTR2B
HTR2C
HTR3A
HTR4
HTR5A
HTR6
HTR7
INSR
LEPR
LEPROT
LEPROTL1
LHCGR
LTB4R
LTB4R2
MC1R
MC2R
MC3R
MC4R
MC5R
MCHR1
MCHR2
MLNR
MTNR1A
MTNR1B
NCOR2
NR0B1
NR0B2
NR1D1
NR1D2
NR1H2
NR1H3
NR1H4
NR1I2
NR1I3
NR2C1
NR2C2
NR2E1
NR2E3
NR2F1
NR2F2
NR2F6
NR3C1
NR3C2
NR4A1
NR4A2
NR4A3
NR5A1
NR5A2
NR6A1
OXER1
OXTR
PGR
POU1F1
PPARA
PPARD
PPARG
PRLHR
PRLR
PTGDR
PTGER1
PTGER2
PTGER3
PTGER4
PTGFR
PTGIR
PTH1R
PTH2R
RARA
RARB
RARG
RORA
RORB
RORC
RXFP1
RXFP2
RXFP3
RXFP4
RXRA
RXRB
RXRG
SCTR
SHBG
SSTR1
SSTR2
SSTR3
SSTR4
SSTR5
TBXA2R
THRA
THRB
TRHR
TSHR
VDR

automatic resolution of annotation disputes

Annotation disputes are automatically resolved after 90 days with the message

If you are happy that the issue raised in your dispute has been addressed, then you need take no further action. If, however, the issue still stands,
then you should contact the group responsible for the annotation....

Can this be prevented?
The dispute should stay open until fixed....
Some fixes take more than a few months to filter through, this is just an additional administrative burden, as this is the only record of the issue.

(After closure the entire thread disappears from view, the thread should really be accessible for future reference , but this is another issue)

AT1G78080

Hi, during my project, I realize that two genes AT1G78080, AT5G47230 are included in GO:0009873 of TAIR data(https://www.arabidopsis.org/index.jsp) but not under the corresponding category of AmiGO2. That confuses me so much. What happens..

how to get all Go terms that in a specific level?

Hi, I am doing GO analysis and I am curious about - if there are any files that contain a list of GO terms in a specific level, for example level 3? Thank you

Querying go-plus.owl with SPARQL in Protege to retrieve GO-UBERON annotations

Hello,

This is partially related to my last issue (#40).

I am trying to do some research about the tissue-specific gene expression (related to genes being more highly expressed in tissues which they effect) and I'm trying to use GO-UBERON (and GO-CL) annotations as part of this work.

For example, I would like to have a list of all GO terms which occur in UBERON tissues (including inferences). I would like not only the GO terms but also which (most specific, i.e. no subclasses of each other) UBERON tissues they occur in.

At the moment, I am able to make some useful DL queries in Protege over the go-plus.owl ontology, but I am unable to export what I need (as I need part of the explanation).

I have heard that a SPARQL query may be more suitable for this task, but while I have been able to write SPARQL queries in Protege for simple ontologies, and I have been able to write simple SPARQL queries that work on the ontobee SPARQL website (http://www.ontobee.org/sparql) over the GO ontology, I have been unable to write any query that works over GO in Protege.

I would appreciate any advice on how to move forwards with this.

Many thanks,

Natalie

Question about logo usage

Hello!

I read the citation policy but still wanted to make sure with you guys that you're happy with me using the GO logo in my Bioconductor sticker logo as can be seen here

My package is primarily designed to run on GO annotation data, which is what my logo is intended to convey. I hope that's all right with you!

Anything I missed, please let me know!

Best regards,
Kevin

EC2GO

Hi,
I am a member of the enzyme commission, and I would like to find out more about who is doing the EC2GO mapping, how often the mapping file is updated, where do you get the latest EC numbers etc. Could you please put me in touch with the responsible person?

Thanks,
Ron Caspi

GO-UBERON annotation queries

I am wondering if there is any guidance for using the annotations between GO and other ontologies (e.g. UBERON), specifically with respect to propagating annotations between two ontologies.

I am trying to do some research about the tissue-specific gene expression (related to genes being more highly expressed in tissues which they effect) and I'm trying to use GO-UBERON annotations as part of this work.

For example, if a given GO biological process term occurs_in an UBERON tissue term, is it safe to say that for example:

• the GO term occurs_in all parents tissues of the UBERON term?
• the GO term occurs_in at least one one of the children of the UBERON term?
• At least one of the children of the GO term occurs_in the UBERON term?
  Etc

And how do these relationships change if the relationship is instead (for example, results_in_morphogenesis_of instead of occurs_in)?

Is there any guidance about this anywhere?

Thanks!

problems with decompress go assocdb-tables.tar.gz file

I have downloaded the latest go-assocdb-tables.tar.gz file for mysql, however, I found that it couldn't work when I use the command 'gunzip' or 'tar' to decompress this file. while I can use the 'tar' to decompress the termdb-tables.tar.gz. So, how I can fix this problems and any useful answers would be appreciated

adjuted P-values are all 1

Hi,

When using GP enrichment analysis I used a set of ExAC constrained genes as my reference set (~10100) and looked for enrichments in my input set of ~500 genes. I got a lot of enrichments > 2 but all my p-values after correction for multiple testing were 1. If there were three genes in the ref set that map to a process and you get all 3 of those in the input set the adjusted p-value will be 1, even though the expected number of genes was 0.1 or something, like and 18 fold enrichment. I'm wondering if this is just not the right kind of test? Or just use all genes in the genome as the ref set because then I get enrichments after corection ofr multuple testing.

Thanks

python code to extract _direct_ protein-protein interactions

Hey,

Is there any easy way to extract direct (Homo sapiens) protein-protein interaction in python code?
I'm struggling as a beginner..

All I need in the columns: The two protein names, interaction type and PMID

Thank you!
Uriel

Discrepancy between AmiGO and remote MySQL-amigo server database

Hi,

I have a question about how AmiGO queries the Gene Ontology database. I have a found a couple instances of AmiGO queries matching gene products to terms that differs from queries I make locally on a copy of the MySQL GO database.

For instance, when I search AmiGO for the gene Herc2 in Mus musculus, 21 annotations are returned. However, when I connect via mysql to the latest version of the remote MySQL database (hosted on mysql-amigo.ebi.ac.uk, with the credentials taken from http://www.geneontology.org/page/lead-database-guide), and load the view "term_J_association_J_gene_product" with the following query:

SELECT * FROM go_latest.term_J_association_J_gene_product
WHERE gp_symbol = 'Herc2'
AND gp_species_id = 336870

I receive 28 annotations. Among these extra 7 annotations are unique terms like "ligase activity" and "ribosomal S6-glutamic acid ligase activity". Even when I download the latest release of the full database, the query results are slightly different.

Another example is Sharpin in Mus musculus, which in my local database is annotated with "nucleus". Via the AmiGO website, nucleus isn't one of the annotations returned for Sharpin.

Is this purely a result of difference in version of the database between the remote mysql-amigo server and the backend database for the AmiGO website? Or are there specific references that are filtered out of queries returned by the AmiGO web interface?

Thank you!
Jason

discovery dates PPI

Hello,

I'm interested in protein-protein interactions. Furthermore, I'm interested in the discovery date of each interaction between each protein couple.

Can I, and how can I extract this data?

Thank you,
Uriel
Technion

Panther tool includes NOT annotations in enrichment

Hello,
I recently noticed that several term enrichment tools are including NOT annotations in their enrichment, Nancy and I between us have contacted g:Profiler, DAVID and ClueGO who are working to remove these annotations. I also tried this on Panther and it seems this too also includes NOT annotations in the enriched set as if they are positive annotations.

I used the protein list copied below. Out of four AGO proteins, only AGO2 (Q9UKV8) is capable of mRNA cleavage and so is annotated to “mRNA cleavage involved in gene silencing by miRNA”. AGO1, 3 and 4 (Q9UL18, Q9H9G7, Q9HCK5) are annotated to the same term but with a NOT qualifier.

I've attached screenshots of the original annotations and the Panther enrichment of this term.

AGO_Annotation.pptx

Q9NRR4
Q8WYQ5
Q9HAV4
P62826
Q9UPY3
Q15633
O75569
Q9UL18
Q9UKV8
Q9H9G7
Q9HCK5

I apologise if this is a known issue, but if we're telling other tools they should ignore NOT annotations then the GOC should also be doing this.

Thanks,
Rachael.

No connection to PANTHER enrichment analysis

I have been unable to access the GO enrichment analysis tool online since mid last week. I am able to browse the site but any search for terms times out with the error that pantherdb.org has no response. It actually looks as though the whole panther site is down. I have been looking on your twitter feed for any service notifications but have seen none. On top of this I am also unable to access your status page. I wonder if this issue is being fixed and whether you know how long it will take to be back up and running?
Thanks!