Open seitza opened 8 years ago
Looks like predictprotein is not configured correctly. Can you please show the contents of your predictproteinrc? Default: /usr/share/predictprotein/predictproteinrc.default
Also, to facilitate the process, please use the example from the snap2 man page (you can simply copy and paste it). It would greatly improve my understanding of the problem if you used the additional debugging option (-d), possibly twice to increase verbosity (-d -d). If the problem persists I would also like to ask you to use the --workdir option to store intermediate files so we can find out whether the blast file really has an unrecognized format.
oh and thanks for pointing out these problems - it shows I need to rework the documentation significantly
@seitza did you find out sth else?
we did not find out something else. we checked the predict protein config, which was wrong per default of course. However, there are some paths, which we could not associate with any of the databases we downloaded and install. furthermore, the documentation of predict protein for standalone tools is short. the man page only states
Resources
Database Cmd line argument
-------- -----------------
big (Uniprot+PDB) blast database --bigblastdb
big_80 (big @ 80% sequence identity --big80blastdb
redunancy level) blast database
swiss blast database --swissblastdb
pfam v2 database --pfam2db
pfam v3 database --pfam3db
prosite_convert.dat --prositeconvdat
Resources for optional targets
Database Cmd line argument
-------- -----------------
big (Uniprot+PDB) blast database --bigblastdb
prosite.dat --prositedat
Swiss-Prot keyword-to-accession --spkeyidx
'index' for loctree
Generating Resources
Courtesy of Wiktor Jurkowski:
* rostlab-data-prosite_convert prosite.dat prosite_convert.dat
* perl /usr/share/loctree/perl/keyindex4loctree.pl < keyindex.txt > keyindex_loctree.txt
* hmmpress Pfam-A.hmm
Until now, we can not handle the generate resources part, because none of the tools comes with the installation process, we came up with by collection infomartion from the readme etc. furthermore, we ignored pram completely, since we are not using it with snap2 (we think at least). the same accounts for prosite.
Running the second example of the snap2 man page, with the following snap2 config file
[snap2]
# snapfun_utildir=path - path to package utilities, default: /usr/share/snap2
snap2dir=/usr/share/snap2
#use snap cache [0|1], default: 0
use_snap_cache=0
#snap cache fetch executable
snapc_fetch=/usr/bin/snapc_fetch
#snap cache store executable
snapc_store=/usr/bin/snapc_store
#snap cache root - overrides snap-cache-mgr configuration
snap_cache_root=
# blastpgp_processors
blastpgp_processors=1
#use predictprotein cache, default: 0
use_pp_cache=0
#predictprotein executable
pp_exe=/usr/bin/predictprotein
#sift executable
sift_exe=/usr/bin/sift_for_submitting_fasta_seq.csh
#reprof executable
reprof_exe=/usr/bin/reprof
#blast executable
blast_exe=/usr/bin/blastpgp
[data]
# swiss_dat=path - location of UniProt/Swiss-Prot dat file
swiss_dat=/home/snap_db/uniprot_sprot.dat
# db_swiss=path - path to ID index of Swiss-Prot dat file (generated by /usr/share/librg-utils-perl/dbSwiss.pl)
db_swiss=/home/snap_db/dbswiss
# PHAT substitution matrix
phat_matrix=/usr/share/snap2/phat.txt
[blast]
# big80=path - path to redundancy reduced database (UniProtKB 80 or equivalent)
big80=/home/snap_db/uniref90.fasta
# swiss=path - path to SwissProt database
swiss=/home/snap_db/uniprot_sprot.fasta
leads to
snap2 -i /usr/share/doc/snap2/examples/MT4_HUMAN.fasta -m all -o MT4_HUMAN.snap2 -d -d --workdir /home/workdir
Working directory: /home/workdir
Job name: MT4_HUMAN
Running external programs..
/usr/bin/predictprotein--seqfile=/home/workdir/MT4_HUMAN.fasta--target=query.blastPsiMat--target=query.psic--target=query.profRdb--target=query.prof1Rdb--target=query.profbval--target=query.isis--target=query.mdisorder--target=query.clustalngz--blast-processors=1--output-dir=/home/workdir at /usr/share/perl5/RG/Snap2/Run.pm line 113
RG::Snap2::Run::predictprotein('/home/workdir', '/home/workdir/MT4_HUMAN.fasta', undef, 1, 'MT4_HUMAN', undef, 0, '/usr/bin/predictprotein', 1, ...) called at /usr/share/perl5/RG/Snap2/Run.pm line 14
RG::Snap2::Run::all('/home/workdir/MT4_HUMAN.allmuts', '/home/workdir', '/usr/bin/sift_for_submitting_fasta_seq.csh', '/home/snap_db/uniref90.fasta', '/home/snap_db/uniprot_sprot.fasta', '/usr/bin/blastpgp', undef, 1, 'MT4_HUMAN', ...) called at /usr/share/perl5/RG/Snap2/Snap2.pm line 223
RG::Snap2::Snap2::run('HASH(0x17bbed8)', 1.0.10) called at /usr/bin/snap2 line 117
Killed
*** ERROR blastpgp_to_saf.pl : *** ERROR blastp_to_saf: blast file format not recognized
Died at /usr/share/librg-utils-perl//blastpgp_to_saf.pl line 76.
make: *** [query.safBlastPsi80] Error 255
make --no-builtin-rules INFILE=query.in -C /tmp/MuvvjepQq6 JOBID=query -j 1 BLASTCORES=1 LIBRGUTILS=/usr/share/librg-utils-perl/ PPROOT=/usr/share/predictprotein/ PROFNUMRESMIN=17 PROFROOT=/usr/share/profphd/prof/ BIGBLASTDB=/home/snap_db/uniref100.fasta BIG80BLASTDB=/home/snap_db/uniref90.fasta PFAM2DB=/mnt/project/rost_db/data/pfam/Pfam_ls PFAM3DB=/mnt/project/rost_db/data/pfam/Pfam-A.hmm PROSITEDAT=/home/snap_db/uniprot_sprot.dat PSICEXE=/usr/share/rost-runpsic/runNewPSIC.pl SPKEYIDX=/mnt/project/rost_db/data/swissprot/keyindex_loctree.txt SWISSBLASTDB=/mnt/project/rost_db/data/blast/swiss --quiet -f /usr/share/predictprotein/MakefilePP.mk query.mdisorder query.profRdb query.prof1Rdb query.profbval query.blastPsiMat query.isis query.clustalngz query.psic failed: 512 at /usr/bin/predictprotein line 391.
Failed to execute '/usr/bin/predictprotein --seqfile=/home/workdir/MT4_HUMAN.fasta --target=query.blastPsiMat --target=query.psic --target=query.profRdb --target=query.prof1Rdb --target=query.profbval --target=query.isis --target=query.mdisorder --target=query.clustalngz --blast-processors=1 --output-dir=/home/workdir': 25 at /usr/share/perl5/RG/Snap2/Run.pm line 114
RG::Snap2::Run::predictprotein('/home/workdir', '/home/workdir/MT4_HUMAN.fasta', undef, 1, 'MT4_HUMAN', undef, 0, '/usr/bin/predictprotein', 1, ...) called at /usr/share/perl5/RG/Snap2/Run.pm line 14
RG::Snap2::Run::all('/home/workdir/MT4_HUMAN.allmuts', '/home/workdir', '/usr/bin/sift_for_submitting_fasta_seq.csh', '/home/snap_db/uniref90.fasta', '/home/snap_db/uniprot_sprot.fasta', '/usr/bin/blastpgp', undef, 1, 'MT4_HUMAN', ...) called at /usr/share/perl5/RG/Snap2/Snap2.pm line 223
RG::Snap2::Snap2::run('HASH(0x17bbed8)', 1.0.10) called at /usr/bin/snap2 line 117
where the predict protein config contains
......
# bigblastdb: Path to comprehensive blast database
# bigblastdb=/mnt/project/rost_db/data/blast/big
bigblastdb=/home/snap_db/uniref100.fasta
# big80blastdb: Path to comprehensive blast database at 80% sequence identity redundancy level
# big80blastdb=/mnt/project/rost_db/data/blast/big_80
big80blastdb=/home/snap_db/uniref90.fasta
# path to profphd
prof=/usr/share/profphd/prof/
# prof numresMin, default: 17
profnumresmin=17
# maximum length of input AA sequence, default: 6000
numresmax=6000
# prositedat: Path to `prosite.dat' file, see L<https://rostlab.org/owiki/index.php/Packages#Resource_definitions>
# prositedat=/mnt/project/rost_db/data/prosite/prosite.dat
prositedat=/home/snap_db/uniprot_sprot.dat
# prositeconvdat: Path to `prosite_convert.dat' file, see L<https://rostlab.org/owiki/index.php/Packages#Resource_definitions>
# prositeconvdat=/mnt/project/rost_db/data/prosite/prosite_convert.dat
prositeconvdat=
# swissblastdb: Path to SwissProt blast database
swissblastdb=/mnt/project/rost_db/data/blast/swiss
# psicexe: psic wrapper executable
psicexe=/usr/share/rost-runpsic/runNewPSIC.pl
# pfam2 db
pfam2db=/mnt/project/rost_db/data/pfam/Pfam_ls
# pfam3 db
pfam3db=/mnt/project/rost_db/data/pfam/Pfam-A.hmm
# Swiss-Prot key 'index' (e.g.) for loctree
spkeyidx=/mnt/project/rost_db/data/swissprot/keyindex_loctree.txt
# default_targets: targets to make if none is given on the command line (separate targets with space)
default_targets=all
EDIT 20.1.16 @hechtm can you have a short look an the files above and propose a method how to get the stuff running on my local machine? There is not tutorial how to install predictportein locally. I guess because it is too labor intensive to do it manually. Of course I found the VMs with preinstalled predictprotein, but in my case I don't really consider them as a solution to my problem. Any ideas
Thanks @seitza
Will look deeper into this this week.
@seitza Hey, just checking all the documentation now. Big sorry for my delay.
Do you have any updates on the tool status? Does it run under normal conditions/parameters?
I've been cleaning a bit the README. Overall you guys have made a GREAT job in documenting everything; it's very clear. Some long detailed information I've moved it to the wiki.
I'm trying now to reproduce everything myself. I've added a few small TODOs in the README. Could you please check? The biggest thing that I'm missing is a clear example of how to run the cli tool. Please write an example and commit possibly necessary sample files in the repo.
I'm fully reachable these days to help resolving these things together.
@seitza @leidigl should we merge branch feature/vagrant
into develop
?
On the same topic, it would be required to add the Vagrantfile
to the develop branch itself.
@jmcejuela It is not necessary to to merge feature/vagrant
into develop
as feature/vagrant
does not include any new data. At the beginning we planned to share the Vagrant file and the .vagrant folder. However this is not possible as different settings for different machines are necessary e.g. the path to the local databases in the Vagrant file. We will delete the branch at the end of this project.
Thanks for the info.
Nonetheless please send me ur file so that I can take a look. Typically vagrantfiles are indeed committable.
U can just drag ur local file to this issue comment and it will get uploaded On Fri, 15 Jan 2016 at 18:43, Linda Leidig notifications@github.com wrote:
@jmcejuela https://github.com/jmcejuela It is not necessary to to merge feature/vagrant into develop as feature/vagrant does not include any new data. At the beginning we planned to share the Vagrant file and the .vagrant folder. However this is not possible as different settings for different machines are necessary e.g. the path to the local databases in the Vagrant file. We will delete the branch at the end of this project.
— Reply to this email directly or view it on GitHub https://github.com/Rostlab/SNAP2/issues/4#issuecomment-172029420.
Hi @jmcejuela here's the Vagrantfile, I use. Please note, that there are some absolute paths, which require to be changed, before it can be used on another machine. Vagrantfile.txt
@seitza thanks André. Concerning this specific runtime issue, does it still happen?
@jmcejuela the runtime issue is not solved yet. I guess, that I just can not configure the databases correctly, because they either are not provided anymore or the updated versions of tools to prepare certain databases return another output than few years ago. I tried it several times, but gave up around christmas.
@seitza thanks for the info.
@GrimSmiler could help on this? They may need the same databases (formats) you use in ConSurf.
I checked the docu of ConSurf and only found references to swissprot and uniprot.
In the config file of SNAP2 there is the following section:
[data]
# swiss_dat=path - location of UniProt/Swiss-Prot dat file
swiss_dat=
# db_swiss=path - path to ID index of Swiss-Prot dat file (generated by /usr/share/librg-utils-perl/dbSwiss.pl)
db_swiss=
# PHAT substitution matrix
phat_matrix=/usr/share/snap2/phat.txt
[blast]
# big80=path - path to redundancy reduced database (UniProtKB 80 or equivalent)
big80=
# swiss=path - path to SwissProt database
swiss=
My problem is, that I cannot find the respective databases or am not sure, which is which, and how to generate the swiss_dat and db_swiss files.
@GrimSmiler successfully installed the DBs, but he still needs to write the documentation for how to get those.
Cross link: https://github.com/Rostlab/ConSurf/issues/9
Status: unresolved
I'm getting a very similar (same?) runtime error as the one reported above. This happens either when I follow the wiki's instructions, and I've even tried to copy the databases from a VM to a local repository, repointing to them, but I'm still getting failures on the blast database format.
mbelmadani@pavdesk-27:~/development/snap2$ snap2 --seq AAAAAAAAAAAAAAAAA --cmdmuts A10S -o ~/test
Working directory: /tmp/lBzgEa74ee
Job name: test
Running external programs..
*** ERROR blastpgp_to_saf.pl : *** ERROR blastp_to_saf: blast file format not recognized
Died at /usr/share/librg-utils-perl//blastpgp_to_saf.pl line 76.
make: *** [query.safBlastPsi80] Error 255
make --no-builtin-rules INFILE=query.in -C /tmp/gqmAMdwEjg JOBID=query -j 1 BLASTCORES=1 LIBRGUTILS=/usr/share/librg-utils-perl/ PPROOT=/usr/share/predictprotein/ PROFNUMRESMIN=17 PROFROOT=/usr/share/profphd/prof/ BIGBLASTDB=/mnt/project/rost_db/data/blast/big BIG80BLASTDB=/mnt/project/rost_db/data/blast/big_80 PFAM2DB=/mnt/project/rost_db/data/pfam/Pfam_ls PFAM3DB=/mnt/project/rost_db/data/pfam/Pfam-A.hmm PROSITEDAT=/mnt/project/rost_db/data/prosite/prosite.dat PROSITECONVDAT=/mnt/project/rost_db/data/prosite/prosite_convert.dat PSICEXE=/usr/share/rost-runpsic/runNewPSIC.pl SPKEYIDX=/mnt/project/rost_db/data/swissprot/keyindex_loctree.txt SWISSBLASTDB=/mnt/project/rost_db/data/blast/swiss --quiet -f /usr/share/predictprotein/MakefilePP.mk query.mdisorder query.clustalngz query.prof1Rdb query.isis query.psic query.profbval query.profRdb query.blastPsiMat failed: 512 at /usr/bin/predictprotein line 391.
Failed to execute '/usr/bin/predictprotein --seqfile=/tmp/lBzgEa74ee/test.fasta --target=query.blastPsiMat --target=query.psic --target=query.profRdb --target=query.prof1Rdb --target=query.profbval --target=query.isis --target=query.mdisorder --target=query.clustalngz --blast-processors=1 --output-dir=/tmp/lBzgEa74ee': 25 at /usr/share/perl5/RG/Snap2/Run.pm line 114.
RG::Snap2::Run::predictprotein('/tmp/lBzgEa74ee', '/tmp/lBzgEa74ee/test.fasta', undef, 1, 'test', undef, 0, '/usr/bin/predictprotein', 0, ...) called at /usr/share/perl5/RG/Snap2/Run.pm line 14
RG::Snap2::Run::all('/tmp/lBzgEa74ee/test.muts', '/tmp/lBzgEa74ee', '/usr/bin/sift_for_submitting_fasta_seq.csh', '/mnt/rostlab-data/big/big_80', '/mnt/rostlab-data/swissprot/uniprot_sprot', '/usr/bin/blastpgp', undef, 1, 'test', ...) called at /usr/share/perl5/RG/Snap2/Snap2.pm line 223
RG::Snap2::Snap2::run('HASH(0x2303f60)', 1.0.10) called at /usr/bin/snap2 line 117
Curiously, I reran the same command immediately and got a different error.
mbelmadani@pavdesk-27:~/development/snap2$ snap2 --seq AAAAAAAAAAAAAAAAA --cmdmuts A10S -o ~/test
Working directory: /tmp/i4lDZ5vM5x
Job name: test
Running external programs..
Incorrect blast file: /tmp/yrCqMGio9J/query.blast at /usr/share/rost-runpsic/runNewPSIC.pl line 384.
main::checkBlast('', 0, '/mnt/project/rost_db/data/blast/big', 'query.fasta', '/tmp/yrCqMGio9J', 'query.blast', 'F', 1) called at /usr/share/rost-runpsic/runNewPSIC.pl line 171
make: *** [query.clustalngz] Error 255
make --no-builtin-rules INFILE=query.in -C /tmp/9ecL7f3R1X JOBID=query -j 1 BLASTCORES=1 LIBRGUTILS=/usr/share/librg-utils-perl/ PPROOT=/usr/share/predictprotein/ PROFNUMRESMIN=17 PROFROOT=/usr/share/profphd/prof/ BIGBLASTDB=/mnt/project/rost_db/data/blast/big BIG80BLASTDB=/mnt/project/rost_db/data/blast/big_80 PFAM2DB=/mnt/project/rost_db/data/pfam/Pfam_ls PFAM3DB=/mnt/project/rost_db/data/pfam/Pfam-A.hmm PROSITEDAT=/mnt/project/rost_db/data/prosite/prosite.dat PROSITECONVDAT=/mnt/project/rost_db/data/prosite/prosite_convert.dat PSICEXE=/usr/share/rost-runpsic/runNewPSIC.pl SPKEYIDX=/mnt/project/rost_db/data/swissprot/keyindex_loctree.txt SWISSBLASTDB=/mnt/project/rost_db/data/blast/swiss --quiet -f /usr/share/predictprotein/MakefilePP.mk query.clustalngz query.isis query.profRdb query.psic query.mdisorder query.profbval query.prof1Rdb query.blastPsiMat failed: 512 at /usr/bin/predictprotein line 391.
Failed to execute '/usr/bin/predictprotein --seqfile=/tmp/i4lDZ5vM5x/test.fasta --target=query.blastPsiMat --target=query.psic --target=query.profRdb --target=query.prof1Rdb --target=query.profbval --target=query.isis --target=query.mdisorder --target=query.clustalngz --blast-processors=1 --output-dir=/tmp/i4lDZ5vM5x': 25 at /usr/share/perl5/RG/Snap2/Run.pm line 114.
RG::Snap2::Run::predictprotein('/tmp/i4lDZ5vM5x', '/tmp/i4lDZ5vM5x/test.fasta', undef, 1, 'test', undef, 0, '/usr/bin/predictprotein', 0, ...) called at /usr/share/perl5/RG/Snap2/Run.pm line 14
RG::Snap2::Run::all('/tmp/i4lDZ5vM5x/test.muts', '/tmp/i4lDZ5vM5x', '/usr/bin/sift_for_submitting_fasta_seq.csh', '/mnt/rostlab-data/big/big_80', '/mnt/rostlab-data/swissprot/uniprot_sprot', '/usr/bin/blastpgp', undef, 1, 'test', ...) called at /usr/share/perl5/RG/Snap2/Snap2.pm line 223
RG::Snap2::Snap2::run('HASH(0x26dfd40)', 1.0.10) called at /usr/bin/snap2 line 117
It seems that if I repeat the command, I randomly get the two different errors. Any suggestions to resolve this?
Thank you @mbelmadani -- May I ask what you use SNAP2 for?
At the moment the tool is not actively maintained (I'm not the author). Your code improvements are very much welcome.
The user @dmitry-n may contribute some fixes in the future
I was interested in looking up the effect of specific point mutations and comparing them with other variant effect prediction scores. I'm aware of snap2web which is nice for a few mutations, but I wanted to do it programmatically.
I ended up using the Biolinux VMWare image. This worked out fine for the time being.
If I have some code improvements, I'll fork and submit a pull requests!
Thanks @mbelmadani - all the best with your project!
Hey everyone,
I was wondering if there was a solution to the blast file format not recognized error. I have followed the steps that are listed here up to "HOWTO workaround the "runtime" error due to probable configuration errors". I can't seem to get that step to work and was wondering if someone could help me. Here is the error that comes up when I try to run the MT4_Human example:
*** ERROR blastpgp_to_saf.pl : *** ERROR blastp_to_saf: blast file format not recognized
Died at /usr/share/librg-utils-perl//blastpgp_to_saf.pl line 76.
make: *** [query.safBlastPsi80] Error 255
make --no-builtin-rules INFILE=query.in -C /tmp/Wis7N3AEN4 JOBID=query -j 1 BLASTCORES=1 LIBRGUTILS=/usr/share/librg-utils-perl/ PPROOT=/usr/share/predictprotein/ PROFNUMRESMIN=8 PROFROOT=/usr/share/profphd/prof/ BIGBLASTDB=/mnt/project/rost_db/data/blast/big BIG80BLASTDB=/mnt/project/rost_db/data/blast/big_80 PFAM2DB=/mnt/project/rost_db/data/pfam/Pfam_ls PFAM3DB=/mnt/project/rost_db/data/pfam/Pfam-A.hmm PROSITEDAT=/mnt/project/rost_db/data/prosite/prosite.dat PROSITECONVDAT=/mnt/project/rost_db/data/prosite/prosite_convert.dat PSICEXE=/usr/share/rost-runpsic/runNewPSIC.pl SPKEYIDX=/mnt/project/rost_db/data/swissprot/keyindex_loctree.txt SWISSBLASTDB=/mnt/project/rost_db/data/blast/swiss --quiet -f /usr/share/predictprotein/MakefilePP.mk query.mdisorder query.profRdb query.prof1Rdb query.profbval query.blastPsiMat query.isis query.clustalngz query.psic failed: 512 at /usr/bin/predictprotein line 391.
Failed to execute '/usr/bin/predictprotein --seqfile=/tmp/K7l6mLm_TT/MT4_HUMAN.fasta --target=query.blastPsiMat --target=query.psic --target=query.profRdb --target=query.prof1Rdb --target=query.profbval --target=query.isis --target=query.mdisorder --target=query.clustalngz --profnumresmin=8 --blast-processors=1 --output-dir=/tmp/K7l6mLm_TT': 25 at /usr/share/perl5/RG/Snap2/Run.pm line 115
RG::Snap2::Run::predictprotein('/tmp/K7l6mLm_TT', '/tmp/K7l6mLm_TT/MT4_HUMAN.fasta', undef, 1, 'MT4_HUMAN', undef, 0, '/usr/bin/predictprotein', 0, ...) called at /usr/share/perl5/RG/Snap2/Run.pm line 14
RG::Snap2::Run::all('/tmp/K7l6mLm_TT/MT4_HUMAN.allmuts', '/tmp/K7l6mLm_TT', '/usr/bin/sift_for_submitting_fasta_seq.csh', '/root/uniref90.fasta', '/root/uniprot_sprot.fasta', '/usr/bin/blastpgp', undef, 1, 'MT4_HUMAN', ...) called at /usr/share/perl5/RG/Snap2/Snap2.pm line 223
RG::Snap2::Snap2::run('HASH(0x2280768)', 1.0.10) called at /usr/bin/snap2 line 117
Any help would be appreciated. Thank you.
I'm very sorry but cannot help with this. Please if you find a solution even if partial that would be really helpful. On Fri 1. Sep 2017 at 18:50, philcharron17 notifications@github.com wrote:
Hey everyone,
I was wondering if there was a solution to the blast file format not recognized error. I have followed the steps that are listed here https://github.com/Rostlab/SNAP2 up to "HOWTO workaround the "runtime" error due to probable configuration errors". I can't seem to get that step to work and was wondering if someone could help me. Here is the error that comes up when I try to run the MT4_Human example:
ERROR blastpgp_to_saf.pl : ERROR blastp_to_saf: blast file format not recognized Died at /usr/share/librg-utils-perl//blastpgp_to_saf.pl line 76. make: *** [query.safBlastPsi80] Error 255
make --no-builtin-rules INFILE=query.in -C /tmp/Wis7N3AEN4 JOBID=query -j 1 BLASTCORES=1 LIBRGUTILS=/usr/share/librg-utils-perl/ PPROOT=/usr/share/predictprotein/ PROFNUMRESMIN=8 PROFROOT=/usr/share/profphd/prof/ BIGBLASTDB=/mnt/project/rost_db/data/blast/big BIG80BLASTDB=/mnt/project/rost_db/data/blast/big_80 PFAM2DB=/mnt/project/rost_db/data/pfam/Pfam_ls PFAM3DB=/mnt/project/rost_db/data/pfam/Pfam-A.hmm PROSITEDAT=/mnt/project/rost_db/data/prosite/prosite.dat PROSITECONVDAT=/mnt/project/rost_db/data/prosite/prosite_convert.dat PSICEXE=/usr/share/rost-runpsic/runNewPSIC.pl SPKEYIDX=/mnt/project/rost_db/data/swissprot/keyindex_loctree.txt SWISSBLASTDB=/mnt/project/rost_db/data/blast/swiss --quiet -f /usr/share/predictprotein/MakefilePP.mk query.mdisorder query.profRdb query.prof1Rdb query.profbval query.blastPsiMat query.isis query.clustalngz query.psic failed: 512 at /usr/bin/predictprotein line 391. Failed to execute '/usr/bin/predictprotein --seqfile=/tmp/K7l6mLm_TT/MT4_HUMAN.fasta --target=query.blastPsiMat --target=query.psic --target=query.profRdb --target=query.prof1Rdb --target=query.profbval --target=query.isis --target=query.mdisorder --target=query.clustalngz --profnumresmin=8 --blast-processors=1 --output-dir=/tmp/K7l6mLm_TT': 25 at /usr/share/perl5/RG/Snap2/Run.pm line 115 RG::Snap2::Run::predictprotein('/tmp/K7l6mLm_TT', '/tmp/K7l6mLm_TT/MT4_HUMAN.fasta', undef, 1, 'MT4_HUMAN', undef, 0, '/usr/bin/predictprotein', 0, ...) called at /usr/share/perl5/RG/Snap2/Run.pm line 14 RG::Snap2::Run::all('/tmp/K7l6mLm_TT/MT4_HUMAN.allmuts', '/tmp/K7l6mLm_TT', '/usr/bin/sift_for_submitting_fasta_seq.csh', '/root/uniref90.fasta', '/root/uniprot_sprot.fasta', '/usr/bin/blastpgp', undef, 1, 'MT4_HUMAN', ...) called at /usr/share/perl5/RG/Snap2/Snap2.pm line 223 RG::Snap2::Snap2::run('HASH(0x2280768)', 1.0.10) called at /usr/bin/snap2 line 117
Any help would be appreciated. Thank you.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/Rostlab/SNAP2/issues/4#issuecomment-326630335, or mute the thread https://github.com/notifications/unsubscribe-auth/AAGQHx_9CLJ52p8Bmb4rQ0T6YUTiNP7Dks5seDW3gaJpZM4Go7A9 .
After finally having installed the tool, now running it fails. First I edited the config file in
~/.snap2rc
:Please note the last lines concerning
$big80
and$swiss
, which I regard as the cause for the error. The rest of the config is default. I tried to enter only the path, the path to the original fast file, a path containing a wildcard to all files, which were generated bymakeblastdb
, and all single files, which were created bymakeblastdb
. All runs failed and result in the following error:A very short personal meeting with @hechtm revealed a a potential problem (using the longest common suffix of all
makeblastdb
-generated files as input path), but didn't help so far. Anyone any ideas?