slagtermaarten / lohhla Goto Github PK

Hello,

LOHHLA is a wonderful tool！
I ran into an issue with running LOHHLA with example data.

My command:
Rscript LOHHLAscript.R
--LOHHLA_loc=/home/shpc_10066/lohhla/
--patientId test
--outputDir /home/shpc_10066/lohhla/example-file/output
--normalBAMfile /home/shpc_10066/lohhla/example-file/bam/example_BS_GL_sorted.bam
--tumorBAMfile /home/shpc_10066/lohhla/example-file/bam/example_tumor_sorted.bam
--BAMDir /home/shpc_10066/lohhla/example-file/bam/
--hlaPath /home/shpc_10066/lohhla/example-file/hlas
--HLAfastaLoc /home/shpc_10066/lohhla/abc_complete.fasta
--CopyNumLoc /home/shpc_10066/lohhla/example-file/solutions.txt
--mappingStep TRUE
--minCoverageFilter 10
--fishingStep TRUE
--cleanUp TRUE
--gatkDir /home/shpc_10066/anaconda3/envs/lohhla/share/picard-2.5.0-2/
--novoDir /home/shpc_10066/anaconda3/pkgs/novoalign-3.07.00-0/bin/
Output was listed in the record file.
record.txt

I'm really at a loss as to how to proceed, and any guidance would be much appreciated!
Thank you for your kind help!

When running this modified script on the example file, I am getting discrepancies in the 'PVal_unique' and 'UnPairedPval_unique' from the solution in the example-file/correct-example-out/example.10.DNA.HLAlossPrediction_CI.xls file. Specifically, the p-values from the modified run are 0.32 and 0.72, respectively, while the original run has p-values of 8e-26 and 1.5e-28. Notably, the 'PVal' and 'UnPairedPval' columns appear similar (but not identical), as do the other columns.

Have the authors encountered a similar issue? I've attached the list of conda packages I have loaded when running LOHHLA and my output.

Thanks!

conda_packages_lohhla.txt

example.10.DNA.HLAlossPrediction_CI.20200204.txt

I ran LOHHLA on my data and got NA values for PVal_unique and UnPairedPval_unique column in the output file. I want to know why this happened and how I can fix it.
I have attached the log file and the output file for your reference. Could you please take a look and let me know what went wrong?

Sample_P17005037LU01.10.DNA.HLAlossPrediction_CI.20230313.txt
Sample_P17005037LU01.10.DNA.IntegerCPN_CI.20230313.txt
running.hla.loh.exome@Mon_Mar_13_06-12-24_2023.log

Hi, I ran the example data with slagtermaarten/LOHHLA and mcgranahanlab/lohhla, and got different results. The PVal_unique=1.01E-24 and UnPairedPval_unique=1.40E-27 from mcgranahanlab/lohhla, but PVal_unique=0.32 and UnPairedPval_unique=0.72 from slagtermaarten/LOHHLA. And they are both differ from the results from correct-example-out. Do you have any idea about that?

Hi, thanks for your excellent software, why does it have NA in copy number?

Dear authors, when running LOHHLA, the following error ocurrs to some of the samples.
I do not understand the warning of "Run with coverageStep completed first".
My code and the warning are pasted bellow, and how can I do to get the correct result.

Rscript /root/lohhla/LOHHLAscript.R --patientId 1218-1 --outputDir /data/1218-1/lohhla-normal/
--normalBAMfile /data/1218-1/1218-2_BS_GL.bam
--BAMDir /data/1218-1/
--hlaPath /data/hla_format/1218-2.hla_format
--HLAfastaLoc /root/lohhla/data/abc_complete.fasta
--CopyNumLoc /data/purity/1218-1.purity
--mappingStep TRUE --minCoverageFilter 10 --fishingStep TRUE --cleanUp FALSE
--HLAexonLoc /root/lohhla/data/hla.dat
--gatkDir /picard --novoDir /novocraft
--coverageStep TRUE

Warning messages:
1: Run with coverageStep completed first -- skipping 1218-1-hla_a!
2: Run with coverageStep completed first -- skipping 1218-1-hla_b!
3: Run with coverageStep completed first -- skipping 1218-1-hla_c!
Error in is.data.frame(x) : object 'combinedTable' not found
Calls: write.table -> is.data.frame
Execution halted

Thank you very much!

Hi! Thanks for your fork provides the option for hg38! But I think there are some bugs in your latest script.

For line 627 and 628, I think they should be removed, otherwise, the command --plottingStep can not be used if the output directory does not exist.

if (plottingStep)
stopifnot(dir.exists(figureDir))

Also, in the plotting parts, some parameters' names are not completed which leads the incomplete of the figure result. (All the plot which xlab = 'HLA genomic position')
For example:
plot(c(1:max(HLA_A_type1normal$V2, HLA_A_type2normal$V2))
, lim = c(0, max(HLA_A_type1normalCov, HLA_A_type2normalCov)), col ='#3182bd99', pch = 16
, lab = 'HLA genomic position'
, lab = 'Coverage'
, ain = c(paste("HLA normal coverage", sample))
, ex = 0.75
, ype = 'n'
, as = 1)