francois-a/fastqtl

Enhanced version of the FastQTL QTL mapper

fastqtl's Issues

how to convert vcf4.2 to vcf4.1?

$ /xxx/software/fastqtl/python/run_FastQTL_threaded.py --help
positional arguments:
vcf Genotypes in VCF 4.1 format
bed Phenotypes in UCSC BED extended format
prefix Prefix for output file name

Hi, I find my vcf file format is ##fileformat=VCFv4.2, but the fastQTL need the vcf4.1.
How can I convert the vcf4.2 to vcf4.1?

Thanks in advance.

The software description page is missing

The documentation for FastQTL can not be found on http://fastqtl.sourceforge.net/. Can you provide new website for the software?

MAF could be the REF clarification

This will look the same clarification that I asked about via the tensorQTL [https://github.com/broadinstitute/tensorqtl/issues/38]!!
But this time I came via the GTEX pipeline documentation page that mentions FastQTL being used.

To triple check, in FastQTL as well as tensorQTL, if the REF is not the major allele in certain cases, then it is reported as the MAF value.

Thanks (again)

issue running fastqtl.static

Hi,

When I was trying to run fastqtl.static, it reports this error message.

terminate called after throwing an instance of 'boost::exception_detail::clone_impl<boost::exception_detail::error_info_injector<std::logic_error> >' what(): character conversion failed Aborted (core dumped)

That's the only error message, it does not show which line throw the error.

Do you have idea on solving this? I could run the original FastQTL.

issue in merge_chunk.py

Hi,

Running run_FastQTL_threaded.py, the log shows that all chunks are finished.

The *_chunk.txt.gz files and log files are generated for all chunks.

The scripts failed at merge_chunks.py step. Here is the screen shot of the error message(because the format can not be reserved when copy&paste), it suggests a syntax error, but apparently the print function call has no syntax error.

I looked into merge_chunk.py I can not find syntax errors, could you have a look at where could be the problem?

Thanks for your time.

BTW, in merge_chunk.py line 43, the help message should be "list of logs" instead of "list of chunks"

cannot find FastQTL in sourceforge.net

Hi,

Thanks for developing this robust tool. I am trying to download this tool, but I can not find it in https://sourceforge.net/. I want to make sure if you have delete it?

Thanks,

Zunpeng

how fastqtl deal with ./. genotype

Hi @francois-a

Thanks for your beautiful code!

I have a question about fastqtl.
How fastqtl deal with ./. genotype?fastqtl impute this site or using other methods to deal ./. genotype?If fastqtl impute this site,how to get the output of impute result .

Thanks in advanced!

run_FastQTL_threaded.py script doen't work

Hello,

We are trying to run the eQTL analysis using this modified version of the FastQTL tool which supports multi-threaded execution. The BED and VCF files we use contain all the chromosomes. We did all the necessary preprocessing steps (replaced all the genotypes different from 0|0, 0|1, 1|0, 1|1 with .|. and the samples are the same in both VCF and BED files). We managed to run successfully the original version of the fastQTL tool on the same files. But we didn't succeed in running this modified script. This is the command we tried to run:

/opt/fastqtl/python/run_FastQTL_threaded.py all_chr.vcf.gz all_genes_expression.bed.gz all_genes --window 1e6 --chunks 100 --threads 16

This is the error we get:
multiprocessing.pool.RemoteTraceback:
"""
Traceback (most recent call last):
File "/usr/lib/python3.5/multiprocessing/pool.py", line 119, in worker
result = (True, func(*args, **kwds))
File "/usr/lib/python3.5/multiprocessing/pool.py", line 44, in mapstar
return list(map(*args))
File "/opt/fastqtl/python/run_FastQTL_threaded.py", line 48, in perm_worker
s = subprocess.check_call(cmd, shell=True, executable='/bin/bash', stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL)
File "/usr/lib/python3.5/subprocess.py", line 581, in check_call
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '/opt/fastqtl/bin/fastQTL --vcf /sbgenomics/Projects/cb8a783b-a338-4a0b-a63b-8d154c5f7ebd/all_test_vcf_tabix.vcf.gz --bed /sbgenomics/Projects/cb8a783b-a338-4a0b-a63b-8d154c5f7ebd/all_genes_expression_tabix.bed.gz --window 1e6 --maf-threshold 0.0 --ma-sample-threshold 0 --chunk 5 100 --out all_genes_expression_tabix_chunk005.txt.gz --log all_genes_expression_tabix_chunk005.log' returned non-zero exit status -9
"""
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/opt/fastqtl/python/run_FastQTL_threaded.py", line 86, in
assert res.get()[0]==0
File "/usr/lib/python3.5/multiprocessing/pool.py", line 608, in get
raise self._value
subprocess.CalledProcessError: Command '/opt/fastqtl/bin/fastQTL --vcf /sbgenomics/Projects/cb8a783b-a338-4a0b-a63b-8d154c5f7ebd/all_test_vcf_tabix.vcf.gz --bed /sbgenomics/Projects/cb8a783b-a338-4a0b-a63b-8d154c5f7ebd/all_genes_expression_tabix.bed.gz --window 1e6 --maf-threshold 0.0 --ma-sample-threshold 0 --chunk 5 100 --out all_genes_expression_tabix_chunk005.txt.gz --log all_genes_expression_tabix_chunk005.log' returned non-zero exit status -9

We also tried to make our own script which runs several tasks in parallel but each time we run this script only one of the tasks finishes successfully, all the others are killed.
We tried everything that came to our mind, but nothing helped. Each time we get the same error. Could you help us to solve this problem?

Thank you in advance,
Best regards,
Veliborka

Bad BED format leading to fastQTL issues

Hello, I have been having trouble making a viable BED file from R that can be recognized by bedtools for sorting.

Error Output: Unexpected file format. Please use tab-delimited BED, GFF, or VCF. Perhaps you have non-integer starts or ends at line 2

I am not sure if I can freely share the file as it contains data from a private database. The file had initial fields (#Chr, start, stop, TargetID). Target ID contains gene names (i.e. APOE) with the other fields having the corresponding information (Chr written as integer). After that, there are over 100 columns containing sample data (expression in decimal notation). The bed file was written with the following R code (On Windows):. As a safety measure, I applied dos2unix to the bed file before using bedtools.

write.table(object, file = "filename", quote = F, sep = "\t", row.names = F, col.names = T)

While bedtools is not recognizing my file, I am able to index it using tabix which is required for fastQTL (https://github.com/francois-a/fastqtl). However, I am unable to read the tabix file with the following error:

Failed to open file "filename.bed.gz.tbi" : Exec format error
Couldn't understand format of "filename.bed.gz.tbi"

The bad tabix format makes in unable for me to use fastQTL, as I get the following error for all chunks used:
Failed to get region 9:37753805-107690518 in [filename.bed.gz]

Coming back to bedtools, it seems to recognize the example file given with the fastQTL repository (examples folder: phenotypes.bed.gz) for the same sort function. As such, I am using bedtools as a type of testing mechanism to see if I am making a valid BED file.

Based on how I have prepared my BED file, are there any issues that would result in improper formatting? Thank you in advance, and please ask if you need any more info. I could probably share the expression data, but I need to check the guidelines of the repository before I do so (AMP-AD consortium).

I did try to switch over to tensorQTL to see if the formatting wasn't as big of an issue with that program, but I am unable to download it from the repository onto my cluster.

--permute parameter only report 1 eQTL result

How to get permute adjust P-val for all eQTL instead of top-eQTL when using --permute 1000 10000? ^_^

genotype match with phenotype

Hello, I am new to this analysis, I would like to know how you can match your patient's genotype with its phenotype, can you please load up the code?

Thanks so much

interaction term

Hello,

To identify potential disease risk eQTLs, we planned to test for an interaction (eQTL x diabetes) term between genotype and diabetes status using the “modelLINEAR_CROSS” argument in MatrixEQTL. But we're wondering if the same type of analysis can be done with FastQTL. Can you please advise?

Cheers,
Ana

Could not make: libRmath.a: No such file or directory

I have an Ubuntu machine, version 18.04.3:

Distributor ID:	Ubuntu
Description:	Ubuntu 18.04.3 LTS
Release:	18.04
Codename:	bionic

After running make I get:

g++ -std=c++11 -O3 obj/analysisNominalBest.o obj/residualizer.o obj/analysisPermutationSequence.o obj/df.o obj/utils.o obj/tabix.o obj/analysisPermutationPerGroup.o obj/mle.o obj/readGenotypes.o obj/readInteractions.o obj/analysisNominalInteractionBest.o obj/analysisPermutation.o obj/analysisNominal.o obj/commands.o obj/readPhenotypes.o obj/analysisNominalInteraction.o obj/fastQTL.o obj/management.o obj/readThresholds.o obj/readInclusionsExclusions.o obj/analysisPermutationInteraction.o obj/analysisMapping.o obj/readGroups.o obj/readCovariates.o lib/Tabix/libtabix.a ../../R-3.2.4/src/nmath/standalone/libRmath.a lib/Tabix/libtabix.a -lm -lz -lbz2 -lboost_iostreams -lboost_program_options -lgslcblas -lgsl -lblas -o bin/fastQTL
g++: error: ../../R-3.2.4/src/nmath/standalone/libRmath.a: No such file or directory
Makefile:74: recipe for target 'bin/fastQTL' failed
make: *** [bin/fastQTL] Error 1

I do have the deb package r-mathlib installed, and it does provide the file libRmath.a, see below:

dpkg-query -L r-mathlib

/.
/usr
/usr/include
/usr/include/Rmath.h
/usr/lib
/usr/lib/libRmath.a
/usr/lib/libRmath.so.1.0.0
/usr/lib/pkgconfig
/usr/lib/pkgconfig/libRmath.pc
/usr/share
/usr/share/doc
/usr/share/doc/r-mathlib
/usr/share/doc/r-mathlib/NEWS.0.gz
/usr/share/doc/r-mathlib/NEWS.1.gz
/usr/share/doc/r-mathlib/NEWS.2.Rd.gz
/usr/share/doc/r-mathlib/NEWS.2.gz
/usr/share/doc/r-mathlib/NEWS.Rd.gz
/usr/share/doc/r-mathlib/NEWS.pdf.gz
/usr/share/doc/r-mathlib/NEWS.rds.gz
/usr/share/doc/r-mathlib/README.gz
/usr/share/doc/r-mathlib/README.mathlib
/usr/share/doc/r-mathlib/changelog.Debian.gz
/usr/share/doc/r-mathlib/changelog.gz
/usr/share/doc/r-mathlib/copyright
/usr/share/doc/r-mathlib/examples
/usr/share/doc/r-mathlib/examples/test.c
/usr/share/lintian
/usr/share/lintian/overrides
/usr/share/lintian/overrides/r-mathlib
/usr/lib/libRmath.so
/usr/lib/libRmath.so.1
/usr/share/doc/r-mathlib/NEWS.gz

Utilizing dosage input to encode a non-genetic 'genotype'

Hello,

I'm attempting to replicate the gtex-pipeline using scripts from the gtex-pipeline repository on GitHub.

However, my use case is a bit different than eQTL mapping. I would like to run a phenotype-QTL mapping, where: instead of many single-variant association tests, I want to test a general phenotype and its association to expression of many genes amongst tissues.

Most importantly, I would like to achieve fast performance on permutation testing for computing p-values of such a phenotype and its association with gene expression (across genes).

My intuition is to encode the phenotype as a dosage for a single genetic variant, across individuals, but I am unsure if this is supported by FastQTL and/or TensorQTL.

So, I'm wondering if this could be possible? If so, could you help me encode this strategy for FastQTL or TensorQTL? I would like to utilize the programs' fast performance in terms of permutation testing.

Thank you

Issue in FastQTL output

Hi
When I used the nominal pass to call sQTLs, I found that there were a lot of repeat sQTLs in the output file named XXX.nominal.XX (example in the below). Can you give me some ideas about this issue?

23:28550049:28588575:clu_28643_NA 23_27774751_A_G -775299 0.0149343 -0.324037
23:28550049:28588575:clu_28643_NA 23_27774805_G_A -775245 7.07646e-06 -0.526193
......
23:28550049:28588575:clu_28643_NA 23_27774751_A_G -775299 0.0149343 -0.324037
23:28550049:28588575:clu_28643_NA 23_27774805_G_A -775245 7.07646e-06 -0.526193
......
23:28550049:28588575:clu_28643_NA 23_27774751_A_G -775299 0.0149343 -0.324037
23:28550049:28588575:clu_28643_NA 23_27774805_G_A -775245 7.07646e-06 -0.526193

Hui

Imputing missing phenotypes

Hi, I have a phenotype data with some missing value. And I found the Fastqtl will impute the issing phenotypes automatically. I wonder if it was imputed by the average value? And how can I avoid imputation?

Error when trying to install FastQTL

Hi,
I've tried to install FastQTL in either my local machine (MacOS) or my VM (ubuntu)

I've installed the required libraries in both system but when trying to install FastQTL by make, they encountered the same errors:

in MacOS:

$ make
g++ -std=c++11 -O3 -D_FAST_CORRELATION -Wall -Wextra -Wno-sign-compare -o obj/readInteractions.o -c src/readInteractions.cpp -Isrc -Ilib/Tabix -Ilib/Eigen -I../../R-3.2.4/src/include/
error: unable to open output file 'obj/readInteractions.o': 'No such file or directory'
1 error generated.
make: *** [obj/readInteractions.o] Error 1

in Ubuntu:

Assembler messages:
Fatal error: can't create obj/readThresholds.o: No such file or directory
Makefile:73: recipe for target 'obj/readThresholds.o' failed
make: *** [obj/readThresholds.o] Error 1

The model in fastQTL.

Hi,

Thanks for developing this robust tool. I am endeavoring to follow in your footsteps by conducting research on eQTL analysis. However, I have encountered some uncertainties concerning the methodology, and I am hopeful that you can provide some suggestions.

I detected many significant signals(25%) in the species I studied. In an effort to address this disparity, I downloaded freely available human data, significant gene-SNP pairs, and generated a Q-Q plot using both human data and our data, limited to significant gene-SNP pairs. The resulting Q-Q plot (following) reveals a noticeable genetic inflation, particularly in the human dataset. Could this inflation be indicative of an underlying population structure that has not been adequately accounted for? Since covariates typically encompass fixed effects, would it be advisable to incorporate random effects, such as a genetic relationship matrix (GRM) or kinship analysis, into our analysis?

Furthermore, I am contemplating the use of GCTA (mlma) for the detection of eQTLs. Do you believe that this approach would be reasonable and potentially address the issues we have encountered in our current analysis?

Thanks

Jun

Output

Hello!
I used fastQTL to conduct an sQTL identification, the command is as follows:
fastQTL.static --vcf merged.vcf.gz --bed Osa_perind.counts.gz.qqnorm_Chr1.gz --cov Osa_perind.counts.gz.PCs.PC5 --permute 1000 10000 --normal --out Osa_perind.permutation. Chr1 --chunk 1 1
This is my first time using permute mode and there are 17 columns of results. I can't understand many columns because they don't have column names. I hope to know the corresponding meanings of these columns.

Compile the fastqtl

Hi Francois,

I am installing fastqtl in the linux and running "./configure" encountered the following errors.

configure: error: --with-x=yes (default) and X11 headers/libs are not available

Thanks in advance!
Best,
Jingjing

An error occurred when I run the script run_fastqtl_threaded.py

I downloaded your repo.
And this is my command :
python run_FastQTL_threaded.py prepare_expression/GTEx_Analysis_2017-06-05_v8_WholeGenomeSeq_866Indiv.vcf.gz prepare_expression/Adipose_Subcutaneous_581_samples__prepare.expression.bed.gz Adipose_Subcutaneous --covariates covariate/Adipose_Subcutaneous_581_samples.PEER_covariates.txt --window 1e6 --chunks 100 --threads 30
Then , a error occurred , this is the whole output information:
[Dec 04 19:17:29] Running FastQTL on 30 threads.
Processing chunk 1
Processing chunk 2
Processing chunk 3
Processing chunk 4
Processing chunk 5
Processing chunk 6
Processing chunk 7
Processing chunk 8
Processing chunk 10
Processing chunk 9
Processing chunk 12
Processing chunk 11
Processing chunk 13
Processing chunk 16
Processing chunk 14
Processing chunk 15
Processing chunk 17
Processing chunk 18
Processing chunk 19
Processing chunk 20
Processing chunk 21
Processing chunk 22
Processing chunk 23
Processing chunk 26
Processing chunk 24
Processing chunk 25
Processing chunk 27
Processing chunk 28
Processing chunk 29
Processing chunk 30
Processing chunk 31
Processing chunk 33
Processing chunk 32
Processing chunk 34
Processing chunk 35
Processing chunk 36
Processing chunk 37
Processing chunk 38
Processing chunk 39
Processing chunk 40
Processing chunk 41
Processing chunk 42
Processing chunk 43
Processing chunk 44
Processing chunk 45
Processing chunk 46
Processing chunk 49
Processing chunk 47
Processing chunk 48
Processing chunk 50
Processing chunk 51
Processing chunk 52
Processing chunk 53
Processing chunk 54
Processing chunk 55
Processing chunk 56
Processing chunk 57
Processing chunk 58
Processing chunk 61
Processing chunk 60
Processing chunk 59
Processing chunk 62
Processing chunk 63
Processing chunk 64
Processing chunk 65
Processing chunk 66
Processing chunk 67
Processing chunk 68
Processing chunk 69
Processing chunk 70
Processing chunk 73
Processing chunk 71
Processing chunk 72
Processing chunk 74
Processing chunk 75
Processing chunk 76
Processing chunk 77
Processing chunk 80
Processing chunk 78
Processing chunk 79
Processing chunk 81
Processing chunk 82
Processing chunk 83
Processing chunk 84
Processing chunk 85
Processing chunk 86
Processing chunk 87
Processing chunk 88
Processing chunk 89
Processing chunk 90
Processing chunk 91
Processing chunk 92
Processing chunk 93
Processing chunk 94
Processing chunk 95
Processing chunk 96
Processing chunk 97
Processing chunk 98
Processing chunk 99
Processing chunk 100
multiprocessing.pool.RemoteTraceback:
"""
Traceback (most recent call last):
File "/picb/lilab2/anaconda3_1/lib/python3.7/multiprocessing/pool.py", line 121, in worker
result = (True, func(*args, **kwds))
File "/picb/lilab2/anaconda3_1/lib/python3.7/multiprocessing/pool.py", line 44, in mapstar
return list(map(*args))
File "/picb/lilab3/wangzengming/fastqtl-master/python/run_FastQTL_threaded.py", line 54, in perm_worker
s = subprocess.check_call(cmd, shell=True, executable='/bin/bash', stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL)
File "/picb/lilab2/anaconda3_1/lib/python3.7/subprocess.py", line 347, in check_call
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '/picb/lilab3/wangzengming/fastqtl-master/bin/fastQTL --vcf prepare_expression/GTEx_Analysis_2017-06-05_v8_WholeGenomeSeq_866Indiv.vcf.gz --bed prepare_expression/Adipose_Subcutaneous_581_samples__prepare.expression.bed.gz --window 1e6 --maf-threshold 0.0 --ma-sample-threshold 0 --interaction-maf-threshold 0 --cov covariate/Adipose_Subcutaneous_581_samples.PEER_covariates.txt --chunk 1 100 --out Adipose_Subcutaneous_chunk001.txt.gz --log Adipose_Subcutaneous_chunk001.log' returned non-zero exit status 127.
"""

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "/picb/lilab3/wangzengming/fastqtl-master/python/run_FastQTL_threaded.py", line 95, in
assert res.get()[0]==0
File "/picb/lilab2/anaconda3_1/lib/python3.7/multiprocessing/pool.py", line 657, in get
raise self._value
subprocess.CalledProcessError: Command '/picb/lilab3/wangzengming/fastqtl-master/bin/fastQTL --vcf prepare_expression/GTEx_Analysis_2017-06-05_v8_WholeGenomeSeq_866Indiv.vcf.gz --bed prepare_expression/Adipose_Subcutaneous_581_samples__prepare.expression.bed.gz --window 1e6 --maf-threshold 0.0 --ma-sample-threshold 0 --interaction-maf-threshold 0 --cov covariate/Adipose_Subcutaneous_581_samples.PEER_covariates.txt --chunk 1 100 --out Adipose_Subcutaneous_chunk001.txt.gz --log Adipose_Subcutaneous_chunk001.log' returned non-zero exit status 127.

Deriving beta-adjusted p-values from nominal p-value and shape parameters

If the number of genotype samples is greater than the phenotype, will genotype samples be used to calculate MAF independently?

Hi，
I get a series of data containing ~80 genotype sample, however ~20 phenotype. I wonder if it is possible to use 80 genotype samples instead of 20 genotype samples to calculate MAF by fastQTL?

fastQTL strand

Hi! I noticed in GTEx expression matrix preparation script, when converting from GTF to BED, for genes on +strand, the coordinates become (start-1, start), and for genes on -strand, coordinates become (end-1, end). I wonder how is FastQTL dealing with strand as the BED matrix does not have a strand column. Is this conversion specific to this extended version of FastQTL, or is it the correct way to prepare expression.bed matrix for general use of FastQTL? Thank you!!

fastQTL no longer normalizes the phenotypes

Hello, I've seen that you removed the option to normalize the phenotypes as was doable by using --normal flag. Was it for a particular motive? Is there any work around? Thank you in advance.

Python exception error with running example data

Hello,
I was trying to run this software with just the basic data included in /example
However, I'm getting the following rather opaque error message - might anyone have any suggestions on how to fix this?

python3 /extra/APPENDICULAR_HEIGHT_PAPER_2021/VARIANT_CALL/fastqtl/python/run_FastQTL_threaded.py genotypes.vcf.gz phenotypes.bed.gz TEST_EXAMPLE --covariates covariates.txt.gz --window 1e6 --ma_sample_threshold 10 --maf_threshold 0.01 --chunks 100 --threads 1 -o test
[Mar 05 19:50:26] Running FastQTL on 1 threads.
Processing chunk 1
Processing chunk 2
Processing chunk 3
Processing chunk 4
Processing chunk 5
Processing chunk 6
Processing chunk 7
Processing chunk 8
Processing chunk 9
Processing chunk 10
Processing chunk 11
Processing chunk 12
Processing chunk 13
Processing chunk 14
Processing chunk 15
Processing chunk 16
Processing chunk 17
Processing chunk 18
Processing chunk 19
Processing chunk 20
Processing chunk 21
Processing chunk 22
Processing chunk 23
Processing chunk 24
Processing chunk 25
Processing chunk 26
Processing chunk 27
Processing chunk 28
Processing chunk 29
Processing chunk 30
Processing chunk 31
Processing chunk 32
Processing chunk 33
Processing chunk 34
Processing chunk 35
Processing chunk 36
Processing chunk 37
Processing chunk 38
Processing chunk 39
Processing chunk 40
Processing chunk 41
Processing chunk 42
Processing chunk 43
Processing chunk 44
Processing chunk 45
Processing chunk 46
Processing chunk 47
Processing chunk 48
Processing chunk 49
Processing chunk 50
Processing chunk 51
Processing chunk 52
Processing chunk 53
Processing chunk 54
Processing chunk 55
Processing chunk 56
Processing chunk 57
Processing chunk 58
Processing chunk 59
Processing chunk 60
Processing chunk 61
Processing chunk 62
Processing chunk 63
Processing chunk 64
Processing chunk 65
Processing chunk 66
Processing chunk 67
Processing chunk 68
Processing chunk 69
Processing chunk 70
Processing chunk 71
Processing chunk 72
Processing chunk 73
Processing chunk 74
Processing chunk 75
Processing chunk 76
Processing chunk 77
Processing chunk 78
Processing chunk 79
Processing chunk 80
Processing chunk 81
Processing chunk 82
Processing chunk 83
Processing chunk 84
Processing chunk 85
Processing chunk 86
Processing chunk 87
Processing chunk 88
Processing chunk 89
Processing chunk 90
Processing chunk 91
Processing chunk 92
Processing chunk 93
Processing chunk 94
Processing chunk 95
Processing chunk 96
Processing chunk 97
Processing chunk 98
Processing chunk 99
Processing chunk 100
multiprocessing.pool.RemoteTraceback:
"""
Traceback (most recent call last):
File "/usr/lib/python3.10/multiprocessing/pool.py", line 125, in worker
result = (True, func(*args, **kwds))
File "/usr/lib/python3.10/multiprocessing/pool.py", line 48, in mapstar
return list(map(*args))
File "/extra/APPENDICULAR_HEIGHT_PAPER_2021/VARIANT_CALL/fastqtl/python/run_FastQTL_threaded.py", line 54, in perm_worker
s = subprocess.check_call(cmd, shell=True, executable='/bin/bash', stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL)
File "/usr/lib/python3.10/subprocess.py", line 369, in check_call
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '/extra/APPENDICULAR_HEIGHT_PAPER_2021/VARIANT_CALL/fastqtl/bin/fastQTL --vcf genotypes.vcf.gz --bed phenotypes.bed.gz --window 1e6 --maf-threshold 0.01 --ma-sample-threshold 10 --interaction-maf-threshold 0 --cov covariates.txt.gz --chunk 1 100 --out TEST_EXAMPLE_chunk001.txt.gz --log TEST_EXAMPLE_chunk001.log' returned non-zero exit status 127.
"""

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "/extra/APPENDICULAR_HEIGHT_PAPER_2021/VARIANT_CALL/fastqtl/python/run_FastQTL_threaded.py", line 95, in
assert res.get()[0]==0
File "/usr/lib/python3.10/multiprocessing/pool.py", line 774, in get
raise self._value
subprocess.CalledProcessError: Command '/extra/APPENDICULAR_HEIGHT_PAPER_2021/VARIANT_CALL/fastqtl/bin/fastQTL --vcf genotypes.vcf.gz --bed phenotypes.bed.gz --window 1e6 --maf-threshold 0.01 --ma-sample-threshold 10 --interaction-maf-threshold 0 --cov covariates.txt.gz --chunk 1 100 --out TEST_EXAMPLE_chunk001.txt.gz --log TEST_EXAMPLE_chunk001.log' returned non-zero exit status 127.

Full column names of fastqtl permutation result.

Hi,
I am confused about the column names of FastQTL result. The introduction of FastQTL (2015) provides 11 columns but the result contains 17 columns. These unknown columns may be important to my downstream analysis. Do you mind providing complete column names?

1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17
PRDM2	129	1.042640	30.89580	11.6750	3.62995e-05	rs2281168	-22037	11	12	0.3157890	-1	0.116541	0.815339	0.305742	0.00199800	0.00082410
CTNNBIP1	121	1.109870	29.46320	12.3000	1.34224e-04	rs946273	-1997	6	6	0.1578950	-1	0.158851	1.564370	0.711198	0.00199800	0.00204789
EMC1	111	0.999246	16.14040	12.1250	2.82725e-04	rs7517930	-14411	9	10	0.2631580	1	0.177539	0.766025	0.374618	0.00199800	0.00457345
ARHGEF10L	203	0.994054	26.04640	12.5250	3.28192e-04	rs7518251	7210	8	9	0.2368420	-1	0.189843	-1.373830	0.702916	0.00899101	0.00877872
PAFAH2	131	1.080160	15.71970	11.5375	1.04571e-03	rs115016376	3876	6	6	0.1578950	-1	0.212256	1.017510	0.562631	0.00899101	0.01134860
NBPF3	164	1.037660	23.27200	11.7625	6.05282e-04	rs12034222	-3170	3	3	0.0789474	1	0.196096	-2.429630	1.271050	0.01198800	0.01173110

Introduction of FastQTL (2015)

1.ID of the tested molecular phenotype (in this particular case, the gene ID)
2.Number of variants tested in cis for this phenotype
3.MLE of the shape1 parameter of the Beta distribution
4.MLE of the shape2 parameter of the Beta distribution
5.Dummy [To be described later]
6.ID of the best variant found for this molecular phenotypes (i.e. with the smallest p-value)
7.Distance between the molecular phenotype - variant pair
8.The nominal p-value of association that quantifies how significant from 0, the regression coefficient is.
9.The slope associated with the nominal p-value of association [only in version > v2-184]
10.A first permutation p-value directly obtained from the permutations with the direct method. This is basically a corrected version of the nominal p-value that accounts for the fact that multiple variants are tested per molecular phenotype.
11.A second permutation p-value obtained via beta approximation. We advice to use this one in any downstream analysis.

Permutation conducted across family members

Hi there, our analysis involves samples from the same family. I was wondering how to conduct permutation taking family structure into account.

One method I can think of is to regress out the relatedness between samples by including family ID as a random effect in the linear mixed effect model (LMM), and then put the residuals got from the LMM model to fastQTL tools and treat them as independent samples and run permutation as usual.

Does this strategy make sense to you, or do you have any other ideas about dealing with relatedness between samples when running permutation with fastQTL? Thank you so much for your help.

All the best,
Diana

The run_FastQTL_threaded.py script produces different output files with the same input files for different chunks number

I am trying to run the run_FastQTL_threaded.py script on the VCF and BED files which contain only chromosome 22 and I noticed that for the different number of chunks the output files are always different. Moreover, as I increase the number of chunks, the output file is smaller. I have checked all output log files and it seems that all the chunks are processed (I expected that some chunks were skipped because of the memory issue but it seems that this is not the case). I am sending you the command lines and the output file's sizes which are produced with these command lines. Please note that these are the command lines that I tried to run locally. But, I tried the same on the AWS m4.16 xlarge instance and the same problem exists there. Moreover, when I run this script on the VCF and BED files which contain all chromosomes, the output files are the same despite the different chunk size between tasks. Could you help me to understand why this is happening when I run the fastQTL tool on chromosome 22?
I used BED file produced by the Geuvadis consortium and VCF file from the 1000 Genomes project.

Number of chunks 10:
Command:
/opt/fastqtl/python/run_FastQTL_threaded.py chr22.vcf.gz FastQTL.bed.gz chunks10 --window 1e6 --chunks 10 --threads 4
Output file size:
chunks10.allpairs.txt.gz - 317 MB

Number of chunks 20:
Command:
/opt/fastqtl/python/run_FastQTL_threaded.py chr22.vcf.gz FastQTL.bed.gz chunks20 --window 1e6 --chunks 20 --threads 4
Output file size:
chunks20.allpairs.txt.gz - 312M

Number of chunks 30:
Command:
/opt/fastqtl/python/run_FastQTL_threaded.py chr22.vcf.gz FastQTL.bed.gz chunks30 --window 1e6 --chunks 30 --threads 4
Output file size:
chunks30.allpairs.txt.gz - 308 MB

Number of chunks 100:
Command:
/opt/fastqtl/python/run_FastQTL_threaded.py chr22.vcf.gz FastQTL.bed.gz chunks100 --window 1e6 --chunks 100 --threads 4
Output file size:
chunks100.allpairs.txt.gz - 265 MB