How to find GO function from GEO2R results

Question

[guntul]

I am too new to RNA sequencing and performed a GEO2R analysis using NCBI website. I obtained some results of down/up regulated genes and values. From this step, how can I find their GO functions? Should I do more analysis or how to access from the analysis results?

GeneID  padj    pvalue  lfcSE   stat    log2FoldChange  baseMean    Symbol  Description
100506497   0.0000309   1.94e-09    0.391   -6.0027842  -2.34933389 13.57   CPVL-AS2    CPVL antisense RNA 2
9242    0.0004013   7.56e-08    0.407   5.3772388   2.18916521  37.51   MSC musculin
84649   0.0004013   5.11e-08    0.405   5.4474355   2.20725151  141.17  DGAT2   diacylglycerol O-acyltransferase 2
101929237   0.0008316   2.09e-07    0.42    -5.1911857  -2.18226839 17.86   LOC101929237    uncharacterized LOC101929237
4099    0.001678    5.27e-07    0.461   5.0161811   2.31151684  414.96  MAG myelin associated glycoprotein
55118   0.0020151   7.60e-07    0.366   4.9454853   1.81171282  351.17  CRTAC1  cartilage acidic protein 1
745 0.0034771   1.75e-06    0.458   4.7806231   2.18905791  84.12   MYRF    myelin regulatory factor
30812   0.0034771   1.53e-06    0.279   4.8071249   1.33943496  104.89  SOX8    SRY-box transcription factor 8
2537    0.0044323   2.51e-06    0.713   4.7076392   3.35635974  5752.59 IFI6    interferon alpha inducible protein 6

Answer 1

[mark.ziemann]

One approach could be to search the gene names on genecards.org and find the GO annotations there. But this can get tedious if you have many significant genes. GO enrichment analysis might be a more systematic approach.

First, you need to set a minimum basemean where genes are classified as detected or not. The list of detected genes is called the background list. Then, after differential expression (DE) analysis, you select the up-regulated and downregulated genes based on significance and fold change thresholds into two separate lists. Go to DAVID (https://david.ncifcrf.gov/) or another web server which does GO enrichment and provide a list of DE genes, along with the correct background list, which will give you your enrichment results.

If you want to do this sort of thing in a more robust way using R, consider this protocol: https://www.protocols.io/view/a-recipe-for-extremely-reproducible-enrichment-ana-j8nlkwpdxl5r/v2

Answer 2

[geneontologyhelp]

You can do an enrichment using PANTHER. Users can perform enrichment analyses directly from the home page of the GOC website, or from https://www.pantherdb.org/. The PANTHER Classification System analysis tool is maintained up to date with GO annotations. The PANTHER classification system is explained in great detail in Mi H et al, PMID: 23868073. The list of supported gene IDs is available from the PANTHER website, but you should be able to use the IDs you have above.

PANTHER also has an API: https://www.pantherdb.org/services/details.jsp

If for some reason you need to convert IDs, we recommend the UniProt mapping tool