addORFfromGTF NA in orfAnalysis columns

[Анастасия Васильева]

Dear Kristofer,

Thank you for the very useful tool and a descriptive manual!

Unfortunately, I don't understand why when I use addORFfromGTF function some known transcripts (that exist in GenCode annotation) have NA in all columns.

Could you kindly help me to understand why this is happening?

Here is my code:

library(IsoformSwitchAnalyzeR)

StringTieCounts <- importIsoformExpression(
    parentDir="ctab", readLength=50
)

myDesign <- data.frame(
    sampleID = colnames(StringTieCounts$abundance)[-1],
    condition = c('control', 'control', 'control', 'knockdown', 'knockdown', 'knockdown')
)

aSwitchList <- importRdata(
    isoformCountMatrix   = StringTieCounts$counts,
    isoformRepExpression = StringTieCounts$abundance,
    designMatrix = myDesign,
    isoformExonAnnoation="merged_annotation.gtf",
    isoformNtFasta ="transcripts.fasta",
    fixStringTieAnnotationProblem = TRUE,
    estimateDifferentialGeneRange = FALSE,
    showProgress = TRUE
    )

FilteredSwitchList = preFilter( aSwitchList )

DEXSeqList <- isoformSwitchTestDEXSeq(FilteredSwitchList)

ORF_known = addORFfromGTF(switchAnalyzeRlist = DEXSeqList,
            pathToGTF = "gencode.v19.chr_patch_hapl_scaff.annotation.gtf")

Thank you in advance for your help!

[Анастасия Васильева]

I wanted to add some information on one of the transcripts, that has all NA columns.

From Gencode annotation (hg19)

chr13    HAVANA    transcript    77575696    77592821    .    -    .gene_id "ENSG00000005812.6"; transcript_id "ENST00000477982.1"; gene_type "protein_coding"; gene_status "KNOWN"; gene_name "FBXL3"; transcript_type "processed_transcript"; transcript_status "KNOWN"; transcript_name "FBXL3-006"; level 2; tag "basic"; havana_gene "OTTHUMG00000017099.3"; havana_transcript "OTTHUMT00000045317.1";

среда, 28 сентября 2022 г. в 14:04:32 UTC+3, Анастасия Васильева:

[Kristoffer Vitting-Seerup]

Did you solve this problem?

Else I can say the one with NA (called MSTAG*) are the once you found with Stringtie so you would need to subsequently use "analyzeNovelIsoformORF()" to analyze those.