I am trying to analyze annotated and novel isoforms, but keep running into problems with the addORFfromGTF step. Mapping was done with HISAT2 and the transcript assembly, merge, and quantification were all done with StringTie. I am using version 1.20.0 of IsoformSwitchAnalyzeR and version 4.2.2. Here is my code up until I get the error:
library(IsoformSwitchAnalyzeR)
library(dplyr)
setwd("C:/Users/Caleb/OneDrive - The Ohio State University/BioinfoData/IsoformSwitch/ENCODE_EFTUD2")
#Import datasets
EFTUD2quant = importIsoformExpression(parentDir = "C:/Users/Caleb/OneDrive - The Ohio State University/BioinfoData/IsoformSwitch/ENCODE_EFTUD2",
addIsofomIdAsColumn = TRUE, showProgress = TRUE, readLength = 100) #reads the t_data.ctab file in each sub directory
names(EFTUD2quant)
#Generate the list of isoform switches
myDesign = data.frame(sampleID = c("SRR4421357","SRR4421358","SRR4422087","SRR4422088"),
condition = c("KD","KD","WT","WT"))
SwitchList = importRdata(isoformCountMatrix = EFTUD2quant$counts,
isoformRepExpression = EFTUD2quant$abundance,
designMatrix = myDesign,
isoformExonAnnoation = "C:/Users/Caleb/OneDrive - The Ohio State University/BioinfoData/IsoformSwitch/ENCODE_EFTUD2/EFTUD2_stringtie_merged.gtf",
showProgress = TRUE)
SwitchList_filt = preFilter(SwitchList,
geneExpressionCutoff = 1, #Cut off genes with less than 1 TPM
isoformExpressionCutoff = 0, #removes unused isoforms
removeSingleIsoformGenes = TRUE, #removes genes with a single isoform
) #The filtering removed 71938 ( 44.11% of ) transcripts. There is now 91136 isoforms left
#Perform switch test (will take a while)
AnalyzedSwitch = isoformSwitchTestDEXSeq(switchAnalyzeRlist = SwitchList_filt, #Uses DEXseq to analyze the isoform switches
alpha = 0.05, #The FDR p-value has to be less than this
dIFcutoff = 0.1, #THere must be at least a 10% change for the isoforms to be considered
reduceToSwitchingGenes = TRUE, #This only keeps genes with one isoform significantly differently used, makes it run faster
showProgress = TRUE)
#Identify ORFs
SwitchAndORF = addORFfromGTF(AnalyzedSwitch,
pathToGTF = "C:/Users/Caleb/OneDrive - The Ohio State University/BioinfoData/IsoformSwitch/Homo_sapiens.GRCh38.108.gtf")
When I run that line it appears to load the GTF fine but I get the following error:
Step 1 of 2: importing GTF (this may take a while)...
Error in `dplyr::full_join()`:
! `...` must be empty.
✖ Problematic argument:
• all = TRUE
When I run rlang::last_error() I get the following:
<error/rlib_error_dots_nonempty>
Error in `dplyr::full_join()`:
! `...` must be empty.
✖ Problematic argument:
• all = TRUE
---
Backtrace:
1. IsoformSwitchAnalyzeR::addORFfromGTF(AnalyzedSwitch, pathToGTF = "C:/Users/Caleb/OneDrive - The Ohio State University/BioinfoData/IsoformSwitch/Homo_sapiens.GRCh38.108.gtf")
4. IsoformSwitchAnalyzeR::importGTF(...)
6. dplyr:::full_join.data.frame(...)
I'm not sure what is going wrong here. I've tried updating all the packages.