GO annotation goana

Question

[2nelly]

Hi all, I am trying to complete a Gene ontology and KEGG pathway analysis through goana and kegga functions of limma. The text below is the filtered output of DE genes from DESeq2

log2 fold change (MLE): condition M vs control 
    Wald test p-value: condition M vs control 
    DataFrame with 1511 rows and 9 columns
                          baseMean     log2FoldChange             lfcSE              stat               pvalue                 padj        symbol  entrezgene              logFC
                         <numeric>          <numeric>         <numeric>         <numeric>            <numeric>            <numeric>      <factor> <character>          <numeric>
    1190007I07Rik 78.4679845005768  -1.17995124143836  0.31699917398998 -3.72225336295565 0.000197452744400775  0.00240695364433439 1190007I07Rik      544717  -1.17995124143836
    1600014C10Rik 3696.58126211061  0.657830466659395  0.09968428410719  6.59913919783016 4.13551890257961e-11 2.76569659480879e-09 1600014C10Rik       72244  0.657830466659395
    1700029J07Rik 133.493553811774 -0.962821564948145 0.246217433451692 -3.91045244624018 9.21233965087002e-05  0.00127090664215766 1700029J07Rik       69479 -0.962821564948145
    1700086O06Rik 93.3960559687328  0.801707051597142 0.281304015805017  2.84996660749001  0.00437238192952888   0.0291164326068455 1700086O06Rik       73516  0.801707051597142
    2310007B03Rik 35.1607046552806   8.26082328924723  1.97467867523721  4.18337595520693 2.87211834203099e-05 0.000481689912787682 2310007B03Rik          NA   8.26082328924723
    ...                        ...                ...               ...               ...                  ...                  ...           ...         ...                ...
    Zfp951        73.3826456533074  -1.70435017720272 0.326546008643756 -5.21932631876716 1.79575124493836e-07 5.60798908054179e-06        Zfp951      626391  -1.70435017720272
    Zfp952        253.747912920015 -0.798965287664422  0.18487890835884 -4.32155995920137  1.5492993698427e-05 0.000276717089305083        Zfp952      240067 -0.798965287664422
    Zfp958        202.398476730763 -0.666889456397734  0.19925444361697   -3.346923884316 0.000817136438920398  0.00767579781184775        Zfp958      233987 -0.666889456397734
    Zfp976        147.014833887861  -0.68891219670236 0.245672538955508  -2.8041888590044  0.00504433480446008   0.0322716136961434        Zfp976      208111  -0.68891219670236
    Zim1           142.21834247356  -1.26769251702809 0.239539774435816 -5.29220051247802 1.20853279701313e-07 3.87636894641962e-06          Zim1       22776  -1.26769251702809

For the annotation I used the entrezIDs of my file as vector and the resulted output seems ok

IDs<-DESeq2_filtered$entrezgene

go <- goana(IDs, species="Mm")

> topGO(go, n=10)
                                                     Term Ont     N   DE         P.DE
GO:0051239 regulation of multicellular organismal process  BP  3239  391 4.769277e-47
GO:0048731                             system development  BP  4726  493 1.748012e-42
GO:0007275             multicellular organism development  BP  5304  522 3.877451e-38
GO:0048856               anatomical structure development  BP  5810  554 4.949658e-37
GO:0032502                          developmental process  BP  6193  580 6.217343e-37
GO:0005515                                protein binding  MF  9005  757 5.128260e-35
GO:0030154                           cell differentiation  BP  4192  431 1.592004e-34
GO:0048869                 cellular developmental process  BP  4417  446 4.934208e-34
GO:0050793            regulation of developmental process  BP  2713  316 5.712812e-34
GO:0005488                                        binding  MF 13344 1006 7.299095e-33

However, I am getting only the DE column and not up and down annotations like the example below, WHICH OF COURSE IS NORMAL since in my vector there is no info regarding up and downregulation

> go <- goana(tr, species="Mm")
> topGO(go, n=15)
                                           Term Ont    N Up Down  P.Up
GO:1903047           mitotic cell cycle process  BP  608  8   84 0.975
GO:0007059               chromosome segregation  BP  277  1   54 0.999
GO:0000070 mitotic sister chromatid segregation  BP  138  0   38 1.000
GO:0000819         sister chromatid segregation  BP  166  0   41 1.000
GO:0022402                   cell cycle process  BP  983 16  108 0.955
GO:0051301                        cell division  BP  524  4   74 0.998
GO:0000278                   mitotic cell cycle  BP  746  9   91 0.992
GO:0000280                     nuclear division  BP  342  6   58 0.815
GO:0098813       nuclear chromosome segregation  BP  219  1   45 0.995
GO:0140014             mitotic nuclear division  BP  239  2   47 0.977
GO:0000776                          kinetochore  CC  126  1   34 0.950
GO:0000775       chromosome, centromeric region  CC  179  1   40 0.986
GO:0007049                           cell cycle  BP 1428 20  132 0.997
GO:0048285                    organelle fission  BP  387  9   58 0.554
GO:0042254                  ribosome biogenesis  BP  266  0   47 1.000
             P.Down
GO:1903047 1.26e-21
GO:0007059 5.21e-21
GO:0000070 1.39e-20
GO:0000819 3.50e-20
GO:0022402 1.01e-19
GO:0051301 1.12e-19
GO:0000278 1.25e-19
GO:0000280 2.08e-19
GO:0098813 1.26e-18
GO:0140014 1.38e-18
GO:0000776 3.02e-18
GO:0000775 4.78e-18
GO:0007049 1.52e-17
GO:0048285 8.20e-17
GO:0042254 1.19e-16

I tried many different ways to pass the log2FC in coef parameter of goana, like using another vector, but unfortunately I didn t manage to get the expected output. Is it possible to create a two lists vector?

I would be grateful if someone has any idea or answer how to pass the log2FC in the GO analysis.

Thank you in advance

Answer 1

[Gordon Smyth]

Just make two vectors of up and down DE genes, then

goana(list(Up=IDs.up,Down=IDs.down), species="Mm")

See ?goana (default method).