there is no t000_.e2e.pdb

[winkingonly]

I run
./run_e2e_ver.sh example/input.fa example/t1

the result:

ll example/t1/

drwxr-xr-x 4 root root 4096 8月 10 17:54 ./
drwxr-xr-x 8 root root 4096 8月 10 16:36 ../
drwxr-xr-x 2 root root 4096 8月 10 17:54 hhblits/
drwxr-xr-x 2 root root 4096 8月 10 17:54 log/
-rw-r--r-- 1 root root 109568 8月 10 17:54 t000_.msa0.a3m
-rw-r--r-- 1 root root 109588 8月 10 17:54 t000_.msa0.ss2.a3m
-rw-r--r-- 1 root root 20 8月 10 17:54 t000_.ss2

there is no t000_.e2e.pdb likes "Expected outputs " said：
For the end-to-end version, there will be a single PDB output having estimated residue-wise CA-lddt at the B-factor column (t000_.e2e.pdb).

why? Thank you!

[IvanWoo22]

I think you should check your logs first.

[winkingonly]

I think you should check your logs first.

logs as follows，there are some error information ,but I also don't konw why and how to deal with it，Can you help me? Thank you!

’log’ folder has six files:
-rw-r--r-- 1 root root 168 8月 10 17:54 hhsearch.stderr
-rw-r--r-- 1 root root 0 8月 10 17:54 hhsearch.stdout
-rw-r--r-- 1 root root 1728 8月 10 17:54 make_msa.stderr
-rw-r--r-- 1 root root 330 8月 10 17:40 make_msa.stdout
-rw-r--r-- 1 root root 368 8月 10 17:54 make_ss.stderr
-rw-r--r-- 1 root root 57 8月 10 17:54 make_ss.stdout

cat hhsearch.stderr

• 17:56:48.618 ERROR: In /opt/conda/conda-bld/hhsuite_1616660820288/work/src/hhalignment.cpp:223: Read:

• 17:56:48.618 ERROR: sequence ss_pred contains no residues.

cat make_msa.stderr

• 17:04:01.600 INFO: Input file = /rosettafold/example/t1/hhblits/t000_.1e-30.a3m

• 17:04:01.600 INFO: Output file = /rosettafold/example/t1/hhblits/t000_.1e-30.id90cov75.a3m

• 17:04:01.626 INFO: Input file = /rosettafold/example/t1/hhblits/t000_.1e-30.a3m

• 17:04:01.626 INFO: Output file = /rosettafold/example/t1/hhblits/t000_.1e-30.id90cov50.a3m

• 17:26:24.387 INFO: Input file = /rosettafold/example/t1/hhblits/t000_.1e-10.a3m

• 17:26:24.387 INFO: Output file = /rosettafold/example/t1/hhblits/t000_.1e-10.id90cov75.a3m

• 17:26:24.433 INFO: Input file = /rosettafold/example/t1/hhblits/t000_.1e-10.a3m

• 17:26:24.433 INFO: Output file = /rosettafold/example/t1/hhblits/t000_.1e-10.id90cov50.a3m

• 17:42:57.979 INFO: Input file = /rosettafold/example/t1/hhblits/t000_.1e-6.a3m

• 17:42:57.979 INFO: Output file = /rosettafold/example/t1/hhblits/t000_.1e-6.id90cov75.a3m

• 17:42:58.017 INFO: Input file = /rosettafold/example/t1/hhblits/t000_.1e-6.a3m

• 17:42:58.017 INFO: Output file = /rosettafold/example/t1/hhblits/t000_.1e-6.id90cov50.a3m

• 17:56:48.030 INFO: Input file = /rosettafold/example/t1/hhblits/t000_.1e-3.a3m

• 17:56:48.031 INFO: Output file = /rosettafold/example/t1/hhblits/t000_.1e-3.id90cov75.a3m

• 17:56:48.071 INFO: Input file = /rosettafold/example/t1/hhblits/t000_.1e-3.a3m

• 17:56:48.072 INFO: Output file = /rosettafold/example/t1/hhblits/t000_.1e-3.id90cov50.a3m

cat make_msa.stdout

hhblits -o /dev/null -mact 0.35 -maxfilt 100000000 -neffmax 20 -cov 25 -cpu 8 -nodiff -realign_max 100000000 -maxseq 1000000 -maxmem 64 -n 4 -d /rosettafold/UniRef30_2020_06/UniRef30_2020_06 -d /rosettafold/bfd/bfd_metaclust_clu_complete_id30_c90_final_seq.sorted_opt
1e-30
1e-10
1e-6
1e-3

cat make_ss.stderr

/rosettafold/input_prep/make_ss.sh: 行 11: /rosettafold/csblast-2.2.3/bin/csbuild: There is no such file or directory
[makemat] FATAL ERROR: Unable to open file t000_.msa0.tmp.chk

cat make_ss.stdout

/root/anaconda3/envs/RoseTTAFold/share/psipred_4.01/data

[IvanWoo22]

Did you run ./install_dependencies.sh?

can be seen from:
/rosettafold/input_prep/make_ss.sh: 行 11: /rosettafold/csblast-2.2.3/bin/csbuild: There is no such file or directory
[makemat] FATAL ERROR: Unable to open file t000_.msa0.tmp.chk

[winkingonly]

Did you run ./install_dependencies.sh?

can be seen from:
/rosettafold/input_prep/make_ss.sh: 行 11: /rosettafold/csblast-2.2.3/bin/csbuild: There is no such file or directory
[makemat] FATAL ERROR: Unable to open file t000_.msa0.tmp.chk

I don't run ./install_dependencies.sh,because I think I run './run_e2e_ver.sh‘，the Readme followed means if run_pyrosetta_ver.sh I should do ./install_dependencies.sh ?

Download and install third-party software if you want to run the entire modeling script (run_pyrosetta_ver.sh)
./install_dependencies.sh

[IvanWoo22]

csblast was installed in ./install_dependencies.sh, so I think you should run it.

[winkingonly]

csblast was installed in ./install_dependencies.sh, so I think you should run it.

thank you I run ./install_dependencies.sh，but there are new errors:
when I run ./run_e2e_ver.sh example/input.fa example/t1

Running HHblits
Running PSIPRED
*** buffer overflow detected ***: psipred terminated
Running hhsearch

make_ss.stderr changed
cat example/t1/log/make_ss.stderr
/rosettafold/input_prep/make_ss.sh: 行 18: 29546Aborted
(core dumped) psipred $ID.mtx $DATADIR/weights.dat $DATADIR/weights.dat2 $DATADIR/weights.dat3 > $ID.ss
Bad psipred pass1 file format!
rm: can't remove '/rosettafold/example/t1/t000_.msa0.a3m.csb.hhblits.ss2': There is no such file or directory

make_ss.stdout changed
cat example/t1/log/make_ss.stdout
/root/anaconda3/envs/RoseTTAFold/share/psipred_4.01/data
Reading CRF from K4000.crf ...
Adding cs-pseudocounts ...
Effective number of sequences exp(entropy) = 11.34
Wrote profile with 138 columns as checkpoint file to t000_.msa0.tmp.chk

[JonathanHungerland]

I have encountered this very same problem while trying to move a working installation from my personal user into a group-accessible environment. I could backtrace that the problem occured in

makemat -P $ID

in the input_prep/make_ss.sh file. My BLASTMAT was still pointing to my user directory and I was able to fix the problem by adding the preceeding line

export BLASTMAT=<your_location>/CondaEnvironments/RoseTTAFold/share/blast-2.2.26/data/
makemat -P $ID

After that the .mtx file that is produced has some integer values of a few hundreds to a few thousand. Before the workaround, the .mtx file consisted of very large negative numbers (most of them identical) which indicated some problem in makemat and caused the buffer overflow in the psipred function that uses the .mtx file. Be aware that make_ss.sh deletes the mtx file at the end so you will need to outcomment the rm statement if you want to check whether this workaround might help for you.

Putting the export statement in one of the parent processes of make_ss.sh did not work for me, so you'd really need to edit make_ss.sh itself.

Edit: this is of course related to #13