Go analysis in OmicsBox (ex Blast2Go)

Question

[ovariohisterectomia]

Hi everyone, i would like to ask for help with this issue. Im really new in bioinformatics. I performed a DE analysis, and i get my gene list for example

GeneID  baseMean    log2FoldChange  lfcSE   stat    pvalue  padj    Protein Name
EGR_00013   64.5598042715165    2.00136220603475    0.231578525192148   8.64226164483152    5.51110427599768e-18    8.66468061170746e-17    hypothetical protein EGR_00013
EGR_05826   185.049060341855    2.00457061223278    0.223024449591272   8.98812043211619    2.5149436651832e-19 4.6174815794538e-18 Focal adhesion kinase
EGR_04778   580.609080320372    2.00638021821051    0.157821469392399   12.7129738807713    5.00925473945179e-37    4.11109536466808e-35    calcium-binding protein

Now i would like to do an GO analysis, and i think my first step should be get the GO terms.

In omicsbox i may see the feature of Functional analysys and BLAST2go anotation.

My questions are:

1. Is that Pipeline right? (get Go terms by go anotation, then perform an enrichment analysys and finally get the charts)

2. How do i load my genes to blast2go to perform this analysis? i tried a .txt file with all columns (geneid, basemean, etc) and even with only geneID but the options doesnt appear for selection...

Please be patient, im a veterinary! thank you!

Answer 1

[jonasmlambert]

Hi.
To perform any kind of functional analysis like e.g. a gene set enrichment analysis or a Fisher's Exact Test you first need the functional annotation of your features (sequence/genes/proteins). These can be found for many organism directly at the Gene Ontology or e.g. the BioMart website.

If you do not have an up-to-date functional annotation of the species you are working with or if there is no straight forward way to link your IDs (EGR_00013, EGR_00013) with any functional information (normally GO terms) then you would need to do a functional annotation yourself. The Blast2GO annotation methodology in OmicsBox (https://www.biobam.com/functional-analysis/) seems the most widely used but there are others too (https://www.nature.com/articles/s41598-018-20211-9/tables/1) In any case, to perform a functional annotation you need the nucleotide or protein sequences for your IDs.

Here a quick start guide and a video tutorial:
- http://docs.blast2go.com/user-manual/quick-start/

Once you have GO terms for your Ids you can run an enrichment analysis. This can be done with online tools, with R or also within OmcisBox where you can import your count table, perform a DE (pairwise/time course) and then directly run different enrichment analysis providing your annotation file (normally a tab separated file with two columns, first the Id and than the GO term, on term per line). If you want to just run the enrichment you would need only your annotations and the list of gene IDs you want to compare with e.g. Up-regulated against the rest.

Fisher´s Exact Test which compares different groups against a reference, e.g. Up-regulated genes against the rest or against down-regulated ones) and GSEA - which finds enrichment functions at the top and button of a ranked list (for example by Fold Change + FDR) without having to define a cutoff. GSEA tutorial:

Hope this helps.