search

pauline-ng / SIFT4G_Create_Genomic_DB

Create genomic databases with SIFT predictions. Input is an organism's genomic DNA (.fa) file and the gene annotation file (.gtf). Output will be a database that can be used with SIFT4G_Annotator.jar to annotate VCF files.
GNU General Public License v3.0
21 stars 7 forks source link

Errors in building my own database for rheMac10 #70

Closed npb596 closed 1 year ago

npb596 commented 1 year ago

Hello,

I have had errors in attempting to create a database for rheMac10 (or MMul_10). There already is a database for rheMac8 (MMul_8) for some reference. I have my own FASTA reference and gtf file and have successfully created a test database for the homo-sapiens_test provided as an examples in test_files. I've allowed the program to run in excess of 24 hours so I think I've gotten the maximal output possible so far.

I'll try my best to describe the errors and give my output and config. The latter should differ only from homo_sapiens-test.txt in the following lines:

PARENT_DIR=/home/npb0015/reinforcement_project/ ORG=macaca_mulatta ORG_VERSION=Mmul_10.99

Running SIFT 4G

SIFT4G_PATH=/tools/sift4g-1/bin/sift4g PROTEIN_DB=/home/npb0015/reinforcement_project/Reference_files/uniprot_sprot.fasta

Sub-directories, don't need to change

GENE_DOWNLOAD_DEST=Reference_files/ CHR_DOWNLOAD_DEST=Reference_files/

The error messages I receive are the following two lines repeated several times:

Use of uninitialized value $aa in string eq at make-single-records-BIOPERL.pl line 274. Use of uninitialized value $aa in concatenation (.) or string at make-single-records-BIOPERL.pl line 280.

Then the following repeated several times:

Use of uninitialized value $orig_aa in string eq at make-single-records-BIOPERL.pl line 551. Use of uninitialized value $mutated_aa in string eq at make-single-records-BIOPERL.pl line 551.

Then the following repeated several times:

Use of uninitialized value in concatenation (.) or string at make-single-records-BIOPERL.pl line 305.

And are are at least a hundred thousand blocks like that. Then later on blocks of the following repeated:

Use of uninitialized value $mutated_aa in string eq at generate-fasta-subst-files-BIOPERL.pl line 894 Use of uninitialized value $aa2 in string eq at generate-fasta-subst-files-BIOPERL.pl line 621. Use of uninitialized value $aa2 in string ne at generate-fasta-subst-files-BIOPERL.pl line 626.

When the program is done it seems to populate singleRecords, subst, and fasta folders with a substantive number of files looking correct. The main folder is populated with log files and the Reference_files directory does have txt files. All other directories in the config file and particularly the Mmul_10.99 is never populated. When I annotated a VCF using the directory it seems as though there's no SIFT annotations occuring.

I hope this has been sufficiently detailed and would greatly appreciate help.

pauline-ng commented 1 year ago

You can ignore the unitialized value messages.

Can you list the folders / files in your sift4g directory ? If there's > 10 files in a folder, just list the first few so I know that there are some files present.

npb596 commented 1 year ago

When you say the sift4g directory I'm assuming you mean the one where I'm calling the sift4g command from? I'm using an HPC that sift4g was loaded to where I can see this directory:

/tools/sift4g-1/

has the folllowing content:

bin LICENSE Makefile README.md sift4g test_files vendor

and inside the sift4g directory above:

LICENSE Makefile obj src

pauline-ng commented 1 year ago

Hi Nick,

I mean list the files that are being generated in the PARENT_DIR set in your config file.

Something like what this user did (scroll down to later messages, where the user lists files & file sizes) https://github.com/pauline-ng/SIFT4G_Create_Genomic_DB/issues/69

npb596 commented 1 year ago

Ok, thank you for the clarification. I'm listing sizes and file names here. I've put a forward slash at the end for directories (anything with 512 is not populated with files).

30M all_prot.fasta 76 fasta.log 277K invalid.log 30M peptide.log 5.8M subst/ 4.0M fasta/ 0 Log2.txt 128K singleRecords/ 8.0K scripts/ 512 SIFT_alignments/ 512 SIFT_predictions/ 512 singleRecords_with_scores/ 512 dbSNP/ 512 Mmul_10.99/

pauline-ng commented 1 year ago

The protein sequences and their substitutions are being generated, but sift4g (actually getting the predictions) isn't being run. This is odd if you are able to run the example files through; then sift4g should run.

What was the last error message on the screen? (Not the unitialized value messages, but the last message). Maybe it ran out of memory?

What happens if you try running sift4g directly: /tools/sift4g-1/bin/sift4g -d /home/npb0015/reinforcement_project/Reference_files/uniprot_sprot.fasta -q all_prot.fasta --subst /home/npb0015/reinforcement_project/subst --out /home/npb0015/reinforcement_project/SIFT_predictions/ --sub-results

npb596 commented 1 year ago

The last messages were:

start siftsharp, getting the alignments /tools/sift4g-1/bin/sift4g -d /home/npb0015/reinforcement_project/Reference_files/uniprot_sprot.fasta -q /home/npb0015/reinforcement_project/ Checking query data and substitutions files

Searching database for candidate sequences

Which makes me think maybe it was running but simply didn't have the time needed to finish... though I gave it 46 hours and 100GB RAM.

Running the command you sent seems to start the process that is being detailed in those messages. Would it be worthwhile to simply run a job using just that command (I'm on an HPC and not wanting to use the login node if that process will take too long)?

pauline-ng commented 1 year ago

Yes, just run the job using that command. After sift4g has completed running, run the script below. This script skips the preceding steps and picks up where it assumes sift4g has finished running.

make-SIFT-db-all-downstream-sift4g.txt

Github doesn't let me upload .pl files, so

1) Rename the file above to make-SIFT-db-all-downstream-sift4g.pl

2) Run script: perl make-SIFT-db-all-downstream-sift4g.pl -conf <config file>

npb596 commented 1 year ago

That seems to be making progress though I do have new issues. Thank you a lot for your help so far. The script I'm running now has the following commands:

/tools/sift4g-1/bin/sift4g -d /home/npb0015/reinforcement_project/Reference_files/uniprot_sprot.fasta -q /home/npb0015/reinforcement_project/all_prot.fasta --subst /home/npb0015/reinforcement_project/subst --out /home/npb0015/reinforcement_project/SIFT_predictions/ --sub-results

perl make-SIFT-db-all-downstream-sift4g.pl -config rheMac10.config.txt

I've uploaded the full output file below. I have tried different variations of this run (e.g. I had files unrelated to SIFT in Reference_files so I had to move those because they were getting gzipped). The Mmul_10.99 directory is populated with appropriate-looking files.

slurm-373314.txt

Given that Mmul_10.99 looks fine to me I did try SIFT4G annotator and received the following errors and the VCF wasn't really annotated at all though. The Mmul_10.99 file, my VCF, and my reference genome are all labeled as chromosomes like "chr1" not "1" so not sure about that error below either.

Chromosome WithSIFT4GAnnotations WithoutSIFT4GAnnotations Progress The following chromosomes (or scaffolds/contigs) are not found in the SIFT 4G database and will not be annotated: 11, 12, 13, 14, 15, 16, 17, 18, 19, 1, 2, 3, 4, 5, 6, 7, 8, 9, 20, 10 Please contact us if you have any questions. ../Mmul_10.99/1.regions does not exist ../Mmul_10.99/2.regions does not exist ../Mmul_10.99/3.regions does not exist ../Mmul_10.99/4.regions does not exist ../Mmul_10.99/5.regions does not exist ../Mmul_10.99/6.regions does not exist ../Mmul_10.99/7.regions does not exist ../Mmul_10.99/8.regions does not exist ../Mmul_10.99/12.regions does not exist ../Mmul_10.99/9.regions does not exist ../Mmul_10.99/14.regions does not exist ../Mmul_10.99/16.regions does not exist ../Mmul_10.99/10.regions does not exist ../Mmul_10.99/15.regions does not exist 12 0 3362 Completed : 1/20 14 0 6878 Completed : 2/20 ../Mmul_10.99/19.regions does not exist ../Mmul_10.99/11.regions does not exist 16 0 939 Completed : 3/20 19 0 895 Completed : 4/20 15 0 2301 Completed : 5/20 ../Mmul_10.99/17.regions does not exist ../Mmul_10.99/13.regions does not exist 17 0 6749 Completed : 6/20 ../Mmul_10.99/18.regions does not exist 11 0 753 Completed : 7/20 10 0 2037 Completed : 8/20 18 0 558 Completed : 9/20 9 0 5513 Completed : 10/20 1 0 1786 Completed : 11/20 ../Mmul_10.99/20.regions does not exist 8 0 2081 Completed : 12/20 5 0 1216 Completed : 13/20 6 0 4514 Completed : 14/20 13 0 2146 Completed : 15/20 2 0 5837 Completed : 16/20 4 0 2636 Completed : 17/20 7 0 1337 Completed : 17/20 3 0 2415 Completed : 19/20 20 0 597 Completed : 20/20

Merging temp files.... SIFT4G Annotation completed ! Output directory:.

pauline-ng commented 1 year ago

  1. Install python3 so that the command python -v shows python3 The *.regions files ars generated by a python script and they need to be present.

  2. Rerun perl make-SIFT-db-all-downstream-sift4g.pl -config rheMac10.config.txt

Don't worry about the chr prefix

npb596 commented 1 year ago

I think python3 is loaded to the HPC by default but to be certain I did load python/3.9.2 into the script and reran. I am receiving the same errors though there are regions files in Mmul_10.99. Mmul_10.99 is produced in reinforcement_project not reinforcement_project/Reference_files as per the config file so is that an issue? I can't guess what else would be right now.

pauline-ng commented 1 year ago

You typepython and it runs python3 and not python2? You did run the test file and it worked right? (the test examples created regions files?)

I don't quite understand your last 2 sentences -- they are unclear and muddled.

pauline-ng commented 1 year ago

According to your first message, your parent directory is :

PARENT_DIR=/home/npb0015/reinforcement_project/

and that's where files are being generated, which is correct.

pauline-ng commented 1 year ago

Please list the files in your singleRecords_with_scores directory

npb596 commented 1 year ago

Apologies for the lack of clarity in my last message. If the files should be in /home/npb0015/reinforcement_project/ then I was just muddling up the issue.

The test example I used did create the relevant regions files and did populate it's own singleRecords_with_scores directory. The singleRecords_with_scores directory for my current attempts with MMul_10.99 is empty. I am definitely running python3, not python2, confirmed with running "python -v".

pauline-ng commented 1 year ago

"The singleRecords_with_scores directory for my current attempts with MMul_10.99 is empty."

This makes me think you don't have SIFT predictions. Can you list directories and their file sizes like you did ~6 days ago?

npb596 commented 1 year ago

30M all_prot.fasta 76 fasta.log 277K invalid.log 30M peptide.log 5.8M subst/ 4.0M fasta/ 0 Log2.txt 128K singleRecords/ 512 SIFT_alignments/ 512 SIFT_predictions/ 128K singleRecords_with_scores/ 512 dbSNP/ 128K Mmul_10.99/

As before, directories with 512 are empty so that suggests the SIFT_predictions directory is still empty. The only two changes in directory size I can see from 6 days ago is that Mmul_10.99 has increased in size (as it is now populated when it wasn't 6 days ago) and singleRecords_with_scores is showing that it's 128K (instead of 512B as before) but it is clear it is actually empty, even searching for hidden files. To be more transparent, reinforcement_project (the directory containing all of the above) does have other directories (e.g. one containing my VCFs) but I assumed it's only worth noting ones being generated by the perl scripts.

pauline-ng commented 1 year ago

Your sift4g isn't finishing if there isn't anything in SIFT_predictions or SIFT_alignments.

pauline-ng commented 1 year ago

Closed due to lack of activity.

npb596 commented 1 year ago

I apologize for the lack of activity, I had issues in my personal life that delayed my work on this for a bit and I realized at some point there was an issue specific to the HPCC I was using, so I was in contact with the HPC admin at my university for a while to troubleshoot that. The issue there specifically is that in the slurm-373314.txt file I posted almost a month ago there was an "Illegal Instruction" error that suggested sift4g was not properly compiled on the amd node I was using. This did not affect my runs with the test_files because I ran those on a login node, which was Intel. The HPC staff graciously recompiled the program on the amd node, so the Illegal Instruction error has been replaced by a segmentation fault error. This is the most recent output I get:

Aligning queries with candidate sequences

This error comes from running the sift4g command given in the error, which is essentially the same as you suggested on October 16th (I get the same error running that command exactly as is). Since a segmentation fault usually means an inability to access the memory required I've tried runs as high as 248GB RAM but still receive the same error, so I assume that is not the issue.

I know this has been a long running issue but I am making progress so I greatly appreciate the help you've provided so far.

pauline-ng commented 1 year ago

Hi Nick,

The issue is the SIFT4G algorithm, which was written by Robert Vaser, a former student in my lab and has a separate github repo. Can you open this issue at this repo: https://github.com/rvaser/sift4g/issues copying the error from the post above?

Also, if you can upload your all_prot.fasta file into dropbox, I can take a look to see if it "looks right"

pauline-ng commented 1 year ago

User removed proteins with X's.

Then got the following error (copied from here):

I removed those proteins, ran sift4g with the resulting FASTA file and there are SIFT prediction files now, thank you. When running the next step:

perl make-SIFT-db-all-downstream-sift4g.pl -config rheMac10.config.txt

I receive the following errors:

Unable to read from /home/npb0015/reinforcement_project/singleRecords/chr11_NW_021160268v1_random.singleRecords_noncoding Traceback (most recent call last): File "/mmfs1/home/npb0015/reinforcement_project/SIFT/scripts_to_build_SIFT_db/make_regions_file.py", line 68, in get_regions (chrom_file, out_file) File "/mmfs1/home/npb0015/reinforcement_project/SIFT/scripts_to_build_SIFT_db/make_regions_file.py", line 31, in get_regions pos = get_pos (first_line) File "/mmfs1/home/npb0015/reinforcement_project/SIFT/scripts_to_build_SIFT_db/make_regions_file.py", line 8, in get_pos return int (fields[0]) ValueError: invalid literal for int() with base 10: '' Unable to read from /home/npb0015/reinforcement_project/singleRecords/chr13_NW_021160278v1_random.singleRecords_noncoding Unable to read from /home/npb0015/reinforcement_project/singleRecords/chr13_NW_021160282v1_random.singleRecords_noncoding Unable to read from /home/npb0015/reinforcement_project/singleRecords/chr18_NW_021160356v1_random.singleRecords_noncoding Unable to read from /home/npb0015/reinforcement_project/singleRecords/chr19_NW_021160375v1_random.singleRecords_noncoding Unable to read from /home/npb0015/reinforcement_project/singleRecords/chr4_NW_021160155v1_random.singleRecords_noncoding Unable to read from /home/npb0015/reinforcement_project/singleRecords/chr6_NW_021160190v1_random.singleRecords_noncoding Unable to read from /home/npb0015/reinforcement_project/singleRecords/chr8_NW_021160222v1_random.singleRecords_noncoding Unable to read from /home/npb0015/reinforcement_project/singleRecords/chrUn_NW_021160452v1.singleRecords_noncoding Unable to read from /home/npb0015/reinforcement_project/singleRecords/chrUn_NW_021160484v1.singleRecords_noncoding Unable to read from /home/npb0015/reinforcement_project/singleRecords/chrUn_NW_021161041v1.singleRecords_noncoding Unable to read from /home/npb0015/reinforcement_project/singleRecords/chrUn_NW_021161561v1.singleRecords_noncoding Unable to read from /home/npb0015/reinforcement_project/singleRecords/chrUn_NW_021161848v1.singleRecords_noncoding Traceback (most recent call last): File "/mmfs1/home/npb0015/reinforcement_project/SIFT/scripts_to_build_SIFT_db/make_regions_file.py", line 68, in get_regions (chrom_file, out_file) File "/mmfs1/home/npb0015/reinforcement_project/SIFT/scripts_to_build_SIFT_db/make_regions_file.py", line 31, in get_regions pos = get_pos (first_line) File "/mmfs1/home/npb0015/reinforcement_project/SIFT/scripts_to_build_SIFT_db/make_regions_file.py", line 8, in get_pos return int (fields[0]) ValueError: invalid literal for int() with base 10: '' Unable to read from /home/npb0015/reinforcement_project/singleRecords/chrUn_NW_021161853v1.singleRecords_noncoding Unable to read from /home/npb0015/reinforcement_project/singleRecords/chrUn_NW_021161876v1.singleRecords_noncoding Unable to read from /home/npb0015/reinforcement_project/singleRecords/chrUn_NW_021162246v1.singleRecords_noncoding Unable to read from /home/npb0015/reinforcement_project/singleRecords/chrUn_NW_021162265v1.singleRecords_noncoding checking the databases zipping up /home/npb0015/reinforcement_project/Reference_files/* gzip: /home/npb0015/reinforcement_project/Reference_files/directory.index.gz already has .gz suffix -- unchanged gzip: /home/npb0015/reinforcement_project/Reference_files/noncoding.txt.gz already has .gz suffix -- unchanged gzip: /home/npb0015/reinforcement_project/Reference_files/protein_coding_genes.txt.gz already has .gz suffix -- unchanged gzip: /home/npb0015/reinforcement_project/Reference_files/rheMac10.ensGene.gtf.gz already has .gz suffix -- unchanged All done!

These appear to be the same errors that I posted here https://github.com/pauline-ng/SIFT4G_Create_Genomic_DB/issues/70 on October 20th, when I still wasn't generating SIFT predictions properly, so not sure why the errors still appear. I am running it with python version 3.9.2. I've looked at the python script a bit and I have some suspicion the ValueError results from a missing or empty chr.gz file but I can't figure what chr would be missing as I know there are chr.gz and regions files.

pauline-ng commented 1 year ago

@npb596

Can you list files & their sizes in: SINGLE_REC_WITH_SIFTSCORE_DIR SINGLE_REC_BY_CHR_DIR

Your gff file, is it all coding? Did it include UTRs?

Thanks, Pauline

npb596 commented 1 year ago

The SINGLE_REC_WITH_SIFTSCORE_DIR is empty. If I recall properly it does populate while running the database script but is empty once finished.

The SINGLE_REC_BY_CHR_DIR directory has 1628 files in it. I will post some representative ones for chr10 (though there are similar files for all chromosomes/scaffolds in the annotation as far as I can tell)

2.9G chr10.singleRecords 1.7G chr10.singleRecords_noncoding 1.4G chr10.singleRecords_noncoding.with_dbSNPid 1.6M chr10.singleRecords_proteins.fa 4.5M chr10.invalidRecords

The gtf file should include just coding transcripts (e.g. no repeat elements or anything) and it does include UTRs as part of that.

pauline-ng commented 1 year ago

@npb596 Please comment out line 74 of make-SIFT-db-all-downstream-sift4g.pl

Old: system ("rm $rm_dir"); New: # system ("rm $rm_dir");

and re-run make-SIFT-db-all-downstream-sift4g.pl

The change will stop the clean up SINGLE_REC_WITH_SIFTSCORE_DIR, and help me debug.

Then send me list & file sizes in SINGLE_REC_WITH_SIFTSCORE_DIR,

npb596 commented 1 year ago

I have done that and now have 987 files in the SINGLE_REC_WITH_SIFTSCORE_DIR. Using chr10 as a representative example again:

0 singleRecords_with_scores/chr10_scores.Srecords.with_dbSNPid.ERR 0 singleRecords_with_scores/chr10_scores.Srecords.with_dbSNPid.LOG 2.9G singleRecords_with_scores/chr10_scores.Srecords.with_dbSNPid.sorted

It seems this pattern follows for all chromosomes/scaffolds. That is, all ERR and LOG files are empty but all the .sorted files have non-zero sizes.

pauline-ng commented 1 year ago

Great, thanks for rerunning it.

Now please run python make_regions_file.py singleRecords_with_scores/chr10_scores.Srecords.with_dbSNPid.sorted tmp.regions

Do you get any errors? Do you see anything in tmp.regions?

npb596 commented 1 year ago

Running that produces the following messages, which don't appear to be errors:

check_SIFTDB.py looking at chr10_scores.Srecords.with_dbSNPid.sorted

and the tmp.regions file is populated with lines like so:

1 699502 OUT 699503 701521 IN 701522 703026 OUT 703027 703263 IN 703264 703285 OUT 703286 704176 IN

alternating OUT and IN until the end of the chromosome.

pauline-ng commented 1 year ago

That's the right output. So I'm stumped.

Please check:

  1. These are non-empty files with read access: singleRecords_with_scores/chr*_scores.Srecords.with_dbSNPid.sorted

(this is input into make-region_files.py)

  1. A folder exists with write-access (this is where the regions file will be written to) $PARENT_DIR/$ORG_VERSION
npb596 commented 1 year ago

Both of those conditions are fulfilled. I may restart the database build from the very beginning just in case there's any changes I made in between the beginning steps to the sift4g step, to the downstream steps, that could affect anything. I can get back to you on how that pans out.

pauline-ng commented 1 year ago

OK, but you'll still have to filter out the fasta sequences with "X" in them.

Can you run: perl make-sift-scores-db-batch.pl $metafile

this calls make_regions_file.py

npb596 commented 1 year ago

I'm assuming there needs to be an actual file in place of $metafile? I don't know which file that is. Assuming I run it exactly that way I get usage statements:

Usage: perl make-sift-scores-db-batch.pl Example: perl make-sift-scores-db-batch.pl mouse.txt

pauline-ng commented 1 year ago

Pass in your config file.

perl make-sift-scores-db-batch.pl <config>

npb596 commented 1 year ago

I am posting the output file from doing that here. 401290-Mmul10db_ds.txt

pauline-ng commented 1 year ago

It's showing no errors. All your regions files should be in:

/home/npb0015/reinforcement_project/Mmul_10.99/

npb596 commented 1 year ago

Yes, there are regions files in that directory. The problem is that running SIFT4G annotator still returns errors and does not annotate. For example, running the following command on a VCF with only chr19:

java -jar SIFT4G_Annotator.jar -c -i Reinforcement.chr19.MAF.ID.vcf -d ../Mmul_10.99/ -r .

will return the following message:

Start Time for SIFT4G code: Tue Dec 06 09:41:29 CST 2022

Started Running ....... Running in Single transcripts mode

Chromosome WithSIFT4GAnnotations WithoutSIFT4GAnnotations Progress The following chromosomes (or scaffolds/contigs) are not found in the SIFT 4G database and will not be annotated: 19 Please contact us if you have any questions. ../Mmul_10.99/19.regions does not exist 19 0 522186 Completed : 1/1

Merging temp files.... SIFT4G Annotation completed ! Output directory:. End Time for parallel code: Tue Dec 06 09:41:39 CST 2022

All the chromosomes are named like "chr19" instead of "19" so I don't understand why it's searching for "19" instead of "chr19". I have also attempted simply copying chr19.regions to a 19.regions in that directory where I can verify it exists using ls and still receive the exact same error. I have also attempted the same thing using the absolute path instead but that doesn't make a difference. I suppose at this point it may just be an error with SIFT4G_annotator then instead of the database generation?

pauline-ng commented 1 year ago

Try a full path for the database

java -jar SIFT4G_Annotator.jar -c -i Reinforcement.chr19.MAF.ID.vcf -d<full path> -r .

npb596 commented 1 year ago

Running the following java -jar SIFT4G_Annotator.jar -c -i Reinforcement.chr19.MAF.ID.vcf -d /home/npb0015/reinforcement_project/Mmul_10.99/ -r .

results in a similar error, even with the copied 19.regions file:

Start Time for SIFT4G code: Tue Dec 06 10:01:37 CST 2022

Started Running ....... Running in Single transcripts mode

Chromosome WithSIFT4GAnnotations WithoutSIFT4GAnnotations Progress The following chromosomes (or scaffolds/contigs) are not found in the SIFT 4G database and will not be annotated: 19 Please contact us if you have any questions. /home/npb0015/reinforcement_project/Mmul_10.99/19.regions does not exist 19 0 522186 Completed : 1/1

Merging temp files.... SIFT4G Annotation completed ! Output directory:. End Time for parallel code: Tue Dec 06 10:01:48 CST 2022

pauline-ng commented 1 year ago

Upload the chr19 files on Dropbox and a VCF file and I can take a look.

npb596 commented 1 year ago

I believe the issue is finally resolved! I mentioned earlier that I had copied "chr19.regions" to "19.regions" and continued to receive the error. However, today I also tried the same thing with chr19.gz -> 19.gz and chr19_SIFTDB_stats.txt -> 19_SIFTDB_stats.txt and now the annotation seems to be working properly. The *SIFT_predictions.vcf now has annotations and they appear to line up with genes I know should be in those locations with the reference genome I'm using.

Thank you so very much for all the help over the past two months, this will hopefully be the last time I'll be bothering you about troubleshooting (though if I have any issues actually using SIFT later I'll be sure to have a different thread opened)!