Citation

When using this resource, please cite: Koopmans et al., 2019, Neuron..

More information about the SynGO consortium can be found on the about page

Geneset tools

input data
The SynGO geneset tool requires your list of gene IDs to either be human hgnc/ensembl/entrez/symbols, mouse MGI or rat RGD. Species other than mouse/rat/human are currently not supported, you should use external tools (eg; biomart) to retreive human orthologs for your gene list for now. You can use our ID conversion tool to map your protein/gene IDs from mouse/rat/human into an ID type supported by SynGO.
background set
The geneset analysis tool intersects your gene list with SynGO annotated genes and compares this to some background set using a Fisher exact test, as detailed here. So picking an appropriate background set is important. We provide a list of all brain expressed genes as a default background set, defined as 'expressed in any GTEx v7 brain tissue'. We encourage the use of custom background lists that are appropriate for your experiment (eg; if you perform an IP against some synaptic gene, you should not be using whole brain as the background set). Don't forget to mention what background set was used when reporting your results.
enrichment analysis
Fisher exact test
For each ontology term, we apply a 1-sided Fisher exact test with 'greater than' hypothesis. The contingency table is built as follows:
your gene list xbackground b
in term tm11 = overlap between x and genes annotated to tm21 = overlap between b and genes annotated to t
not in term tm12 = total size of x (including genes not mapped to SynGO DB) - m11m22 = total size of b (including genes not mapped to SynGO DB) - m21
finding enriched terms
To find enriched terms within the entire SynGO ontology, we first select the most specific term where each 'gene cluster' (unique set of genes) is found and then apply multiple testing correction using False Discovery Rate (FDR) on the subset of terms that contain these 'gene clusters'. The result is found in the "q-value" column of the output tables on the website and Excel files availble through 'Download analysis' (NaN entries indicate the term either holds too few genes or was not a 'gene cluster')

SynGO annotation systematics

Evidence tracking

To systematically track evidence, classifications were designed for the model systems used. For synaptic localization (CC), microscopy and biochemical studies were defined as the main experimental classes, each with several sub-classes. For functional studies, experimental classes were defined based on perturbation type and the methodology (assay) used to detect the consequences, again with several sub-classes. These classifications were made coherent with the Evidence and Conclusions Ontology (ECO) (Giglio et al., 2018), and new ECO terms were defined. Together, these three dimensions of evidence, (i) model system/preparation, (ii) experimental perturbation and (iii) assay, provide a systematic, coherent and detailed definition of the evidence to annotate synaptic genes.

Detailed reference to these three dimensions of evidence was stored as part of each annotation (PubMed ID, figure numbers, panels), providing a detailed rationale for each annotation, which can be reviewed by SynGO users.

SynGO paper Figure S2

Types of membrane association

We follow GO-principles in distinguishing different types of membrane association using the following terminology

membrane association types

  • intrinsic component of membrane
    proteins that have some covalently attached moiety embedded in the membrane
    • integral component of membrane
      proteins in which some part of the peptide sequence spans all or part of the membrane. A subclass of this covers transmembrane proteins; those that completely span the membrane
    • anchored component of membrane
      Proteins tethered to a membrane by a covalently attached anchor, such as a lipid moiety, which is embedded in the membrane
  • extrinsic component of membrane
    proteins that are associated with membranes, but are neither directly embedded in the membrane nor anchored by covalent bonds to any moiety embedded in the membrane. Proteins that associate indirectly via protein-protein interactions are not annotated as membrane-associated protein by SynGO (see examples below)
To illustrate the use of membrane terminology we provide examples of several known membrane-associated proteins:
namefeatureannotation
GluR1has transmembrane domainintegral component of postsynaptic specialization membrane
PSD-95has functional palmitoylation siteanchored to postsynaptic specialization membrane
PKCɑhas lipid-binding C2 domainextrinsic component of postsynaptic membrane

Submit your data

The SynGO database is ever growing and thriving on community participation. We welcome suggestions for synaptic location/function annotations that are currently missing, head over to the submit page to contribute to this project.

Technical support

supported browsers
This website works best with Chrome, Firefox and Safari. Javascript must be enabled. We've also tested Microsoft Edge and various Chromium-based browsers. So far, only the (ancient) Internet Explorer browser and it's predecessors has been found incompatible.
working with SVG files
The SynGO tools export graphics as SVG files (a vector graphics format) so you can incorporate these as high quality figures into your presentations and publications. You can use your browser as a viewer by simply drag&drop'ing the downloaded SVG onto your browser. SVGs can be edited using e.g. Adobe Illustrator or LibreOffice Draw.

Privacy Policy

ID mapping tool
For user convenience we built an ID mapping tool to converts protein/gene IDs to a format compatible with our database. The external webservice MyGene.info performs the actual conversion of your input data. Your data is directly sent to this webservice from your browser, your data does not pass through the SynGO webserver at any point. Privacy terms of the external website apply for this tool, see further MyGene.info terms of use.
The SynGO geneset analysis tool does not store your data
SynGO data is downloaded and cached in your computer's local storage for efficiency upon first loading this website. This enables the geneset analysis to be performed locally within your browser using the pre-downloaded SynGO data and javascript code. So your input data for geneset analysis does not leave your computer, is not stored anywhere else and remains private at all times.
Cookies and Google Analytics
We use the minimal set of Google Analytics functions to count how many visitors each page on this website receives. This allows us to keep tabs on the website's usage stats, eg; how many hits does the help page get? Your IP address is anonymized in the process of counting page visits. Your gene lists are not stored or shared anywhere, as detailed in the above Privacy Policy section.

Please whitelist this website in your ad blocker; we do not serve any advertisements and only track page counts which, together with citations of our paper(s), is a key metric for demonstrating the popularity of this data resource in our scientific community!

Contact