DEXSeq error

[jalalsiddiqui]

I tried to run it with the DEXSeq parameter and got the following error.

Step 1 of 3 : Detecting isoform switches...

error: matrix multiplication: incompatible matrix dimensions: 56x30 and 0x1 Error in fitBeta(ySEXP = ySEXP, xSEXP = xSEXP, nfSEXP = nfSEXP, alpha_hatSEXP = alpha_hatSEXP, : matrix multiplication: incompatible matrix dimensions: 56x30 and 0x1 Calls: isoformSwitchAnalysisCombined ... tryCatch -> tryCatchList -> tryCatchOne ->

[kvittingseerup]

That does not look right. Could you run traceback() right after you get this error and post the resutl. Could you also post your design matrix?

[jalalsiddiqui]

The info is as follows:

PART 1: EXTRACTING ISOFORM SWTICH SEQUENCES Step 1 of 3 : Detecting isoform switches...

error: matrix multiplication: incompatible matrix dimensions: 56x30 and 0x1 Error in fitBeta(ySEXP = ySEXP, xSEXP = xSEXP, nfSEXP = nfSEXP, alpha_hatSEXP = alpha_hatSEXP, : matrix multiplication: incompatible matrix dimensions: 56x30 and 0x1 Calls: isoformSwitchAnalysisCombined ... tryCatch -> tryCatchList -> tryCatchOne -> In addition: Warning message: In isoformSwitchTestDEXSeq(switchAnalyzeRlist, reduceToSwitchingGenes = TRUE, : Most journals and scientists consider an alpha larger than 0.05 untrustworthy. We therefore recommend using alpha values smaller than or queal to 0.05 Execution halted

The design matrix is

[1] "list_allsamples" "Group2" "Group1" sampleID condition SRR01 SRR01 Group2 SRR02 SRR02 Group1 SRR03 SRR03 Group1 SRR04 SRR04 Group1 SRR05 SRR05 Group1 SRR06 SRR06 Group2 SRR07 SRR07 Group1 SRR08 SRR08 Group1 SRR09 SRR09 Group2 SRR10 SRR10 Group2 SRR11 SRR11 Group1 SRR12 SRR12 Group2 SRR13 SRR13 Group1 SRR14 SRR14 Group2 SRR15 SRR15 Group1 SRR16 SRR16 Group2 SRR17 SRR17 Group2 SRR18 SRR18 Group2 SRR19 SRR19 Group2 SRR20 SRR20 Group1 SRR21 SRR21 Group1 SRR22 SRR22 Group1 SRR23 SRR23 Group2 SRR24 SRR24 Group2 SRR25 SRR25 Group2 SRR26 SRR26 Group1 SRR27 SRR27 Group1 SRR28 SRR28 Group1

[kvittingseerup]

Does your design matrix have 2 or 3 colums? Where does the [1] "list_allsamples" "Group2" "Group1" output come from?

Also could you send me the output of running:

summary(<your_switchAnalyzeRlist>)

And which version of IsoformSwitchAnalyzeR you are running? You can test that with this comand: packageVersion('IsoformSwitchAnalyzeR')

[jalalsiddiqui]

The design matrix has 2 columns. list_all_samples is basically the list of the samples. Group1 and Group2 are the 2 groups compared. The two columns are sampleID and condition.

The info needed is as follows.

This switchAnalyzeRlist list contains: 87852 isoforms from 14471 genes 1 comparison from 2 conditions (in total 28 samples)

Feature analyzed: [1] "ORFs, ntSequence" [1] ‘1.11.3’

Let me know if this is what you need.

[kvittingseerup]

That all looks right. Could you try restarting R and trying again?

Which version of DEXSeq and DESeq2 are you using?

[jalalsiddiqui]

DEXSeq_1.28.3, DESeq2_1.22.2

[kvittingseerup]

Hi Jalal

Could you try updating DEXSeq and DESeq2 to the newest version from: https://bioconductor.org/packages/release/bioc/html/DESeq2.html https://bioconductor.org/packages/release/bioc/html/DEXSeq.html

Remember to restart R before re-running the IsoformSwitchAnalyzeR analysis (necessary before the update take effects).

Cheers Kristoffer

Den tir. 16. jun. 2020 kl. 22.24 skrev Jalal K. Siddiqui < notifications@github.com>:

DEXSeq_1.28.3, DESeq2_1.22.2

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/kvittingseerup/IsoformSwitchAnalyzeR/issues/69#issuecomment-644992690, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFU7JY4VZG6CKGA57N73DWDRW7IH3ANCNFSM4N2Y3LJA .

[jalalsiddiqui]

Sorry about late reply. I can do that. Do you think it is best that use the latest version of R as well?

[kvittingseerup]

R 3.6.1 or larger should do it :-)

[jalalsiddiqui]

I am still working on trying to get this to install.

The downloaded source packages are in ‘/tmp/RtmpzmfxWi/downloaded_packages’

BiocManager::install("IsoformSwitchAnalyzeR") Error in get(method, envir = home) : lazy-load database '/users/PAS1143/osu8165/R/x86_64-pc-linux-gnu-library/3.6/BiocManager/R/BiocManager.rdb' is corrupt In addition: Warning messages: 1: In .registerS3method(fin[i, 1], fin[i, 2], fin[i, 3], fin[i, 4], : restarting interrupted promise evaluation 2: In get(method, envir = home) : restarting interrupted promise evaluation 3: In get(method, envir = home) : internal error -3 in R_decompress1 library(IsoformSwitchAnalyzeR) Error in library(IsoformSwitchAnalyzeR) : there is no package called ‘IsoformSwitchAnalyzeR’

I am trying to install on R version 3.6.3.

This may be an issue on the end of the cluster.

[jalalsiddiqui]

I reinstalled the package on R 4.0 on my laptop and got the following results.

PART 1: EXTRACTING ISOFORM SWTICH SEQUENCES Step 1 of 3 : Detecting isoform switches... Step 3 of 3 : Extracting (and outputting) sequences The 'removeLongAAseq' and 'removeShortAAseq' arguments: Removed : 0 isoforms. Trimmed : 0 isoforms (to only contain the first 1000 AA)

The number of isoform switches found were: Comparison nrIsoforms nrSwitches nrGenes 1 Non_responder vs Responder 5 6 5 The nucleotide and amino acid sequences of these isoforms have been outputted to the supplied directory. These sequences enabling external analysis of protein domians (Pfam), coding potential (CPAT/CPC2) or signal peptides (SignalIP). See ?analyzeCPAT, ?analyzeCPC2, ?analyzePFAM or?analyzeSignalIP (under details) for suggested ways of running these three tools.

PART 2: PLOTTING ISOFORM SEQUENCES Step 1 of 5 : Importing external sequence analysis... Step 2 of 5 : Analyzing alternative splicing... Step 3 of 5 : Prediciting functional consequences... Step 4 of 5 : Making indidual isoform switch plots... |=================================================================================================| 100% Step 5 of 5 : Analyzing combined consequences plot... Error in extractConsequenceEnrichment(switchAnalyzeRlist = switchAnalyzeRlist, : No features left for ploting after filtering with via "minEventsForPlotting" argument. In addition: There were 11 warnings (use warnings() to see them)

traceback() 4: stop("No features left for ploting after filtering with via \"minEventsForPlotting\" argument.") 3: extractConsequenceEnrichment(switchAnalyzeRlist = switchAnalyzeRlist, alpha = alpha, dIFcutoff = dIFcutoff, plot = TRUE, returnResult = FALSE) 2: isoformSwitchAnalysisPart2(switchAnalyzeRlist = switchAnalyzeRlist, alpha = alpha, dIFcutoff = dIFcutoff, n = n, consequencesToAnalyze = consequencesToAnalyze, pathToOutput = pathToOutput, fileType = fileType, asFractionTotal = asFractionTotal, outputPlots = outputPlots, quiet = quiet) 1: isoformSwitchAnalysisCombined(switchAnalyzeRlist = aSwitchListFiltered, alpha = 0.2, dIFcutoff = 0.1, switchTestMethod = "DEXSeq", n = NA, pathToOutput = "combinedDEXSeq_isar_results", overwriteORF = FALSE, outputSequences = TRUE, genomeObject, orfMethod = "longest", cds = NULL, consequencesToAnalyze = c("intron_retention", "ORF_seq_similarity", "NMD_status"), fileType = "pdf", asFractionTotal = FALSE, outputPlots = TRUE, quiet = FALSE)

[kvittingseerup]

Good that you solved the installation/dependencies problem.

The latest problem is that you "only" find 5 genes with isoform switches and the extractConsequenceEnrichment() function requires 10 genes per default.

Thanks for pointing this out - I will fix it in a future update. In the meantime you can follow the detailed workflow - specifically you can use the switchAnalyzeRlist resulting from running isoformSwitchAnalysisPart1() and start from this step of the detailed workflow.

Cheers Kristoffer

[jalalsiddiqui]

Thanks for all your help. Greatly appreciate it. Sorry for the delayed reply