Sequence Matrix: loading file into memory

[qiwill]

hi, I first time to use Sequence Matrix and Species Identifier,but the program will say "loading XXXfile into memory.Sorry for the delay!" when I loading .fasta file. I had tried to change -Xmx1000M to -Xmx2000M, but it still error. My computer have 12Gb RAM. How to solve this problem? THX!

[gaurav]

Hi @qiwill! Hm, that should work. How big is your FASTA file? Could you please e-mail me a copy of your FASTA file and I'll try to get it working on my end? My e-mail address is in my Github profile and on my website (http://ggvaidya.com/)

[AlejandraPanzera]

Hello! I have the same problem as qiwill. Will you able to solve it? My files are 3 to 7 KB each. Thank you very much in advanced!

[gaurav]

I'll give it a shot -- please e-mail me your files!

[gaurav]

(You can also attach them here if you'd like!)

[qiwill]

Hi, @AlejandraPanzera Your sequences have been aligned before you load it to SpeciesIdentifier?

[CLMonteil]

Hi all, I have the same issues, setting -Xms and -Xmx does not change anything into this issue...

[CLMonteil]

I just would have to add an observation. I have 32 GB of RAM, i m not limited. If I want to concatenate 1000 proteins sequences, the program cannot achieve the task (out of memory) if I do not replace external gaps by "?". If I do, then No problem.

[gaurav]

Thanks for the report, @CLMonteil, and that is a eally interesting observation! Could you please e-mail me some of the files in which you're seeing this behavior? Which operating system are you seeing this problem in?

[CLMonteil]

Hi Gaurav,

What I told you yesterday was true but… yesterday. I tried again with the same dataset today, and results were pretty inconsistent. I mean that I opened the terminal (MacOS HS 3.2 GHz i5, 32G memory)

I wrote the following command: java -jar -Xms10000m -Xmx12000m /Applications/SequenceMatrix/Sequence\ Matrix.app/Contents/Resources/Java/SequenceMatrix.jar

Then I added the 836 files to it, then I said « not at all ». Everything ran perfectly. I got a concatenation! I did quit the terminal and sequencematrix. I performed again and again the same procedure accepting or not to change external gaps with « ? », I did not work while nothing has changed and no other application was opened.

So I really do not know why the same procedure works sometimes and not most of the time... I sent some data to you. some files do not have the same taxa.

Thank you,

Caroline ——

Caroline MONTEIL, PhD

Laboratoire de Bioénergétique Cellulairehttp://biam.cea.fr/drf/biam/Pages/laboratoires/lbc.aspx Institut de Biosciences et Biotechnologies d’Aix-Marseille (BIAM) UMR 7265 CNRS-CEA-Université Aix-Marseille II CEA Cadarache, Bât. 156, 13108 ST PAUL LEZ DURANCE Tel: +33 (0)4 42 25 46 19 Mail [email protected]mailto:[email protected] Researchgatehttps://www.researchgate.net/profile/Caroline_Monteil

[cid:[email protected]][cid:[email protected]]

Le 2 août 2018 à 19:57, Gaurav Vaidya <[email protected]mailto:[email protected]> a écrit :

Thanks for the report, @CLMonteilhttps://github.com/CLMonteil, and that is a eally interesting observation! Could you please e-mail me some of the files in which you're seeing this behavior? Which operating system are you seeing this problem in?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/gaurav/taxondna/issues/63#issuecomment-410014611, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AoFo402C-09jSUk3_9sBnjorca_Op9puks5uMz2lgaJpZM4JTjz7.

[willyard]

I am trying Sequence Matrix for the first time on my Windows 7 Professional PC with 16 GB. I need to concatenate 447 gene alignments that have up to 81 samples each. The files are fairly small (under 50 KB each), but not all of them have all of the samples. I was excited to see the first few genes load into the matrix and that this useful software can handle the alignments with missing samples. But the program always hangs up with a "loading .... into memory" error. This happens on the 3rd or the 6th or the 7th sequence. I am still running the .jar file because I cannot gett the .bat file to run. I understand that I have not yet run it with the memory changed to -Xmx2000M. I think this problem running the .bat just has to do with how %windir% works to find the .jar - it has been a long time since I needed DOS. But based on the discussions in this thread, I wonder if I will have any chance assembling this size matrix. Maybe this is too big a matrix for your java-based program. Do you have any suggestions? Should I keep trying to run it with more memory? Thank you so much, Ann Willyard

[CLMonteil]

Hi all,

It looks like if I quit any other software already opened, SequenceMatrix may handle the job even if taxa number is different per file if I increase the memory. So increasing the memory may not work if some other software is running or even sleeping.

best

Caroline

——

Caroline MONTEIL, PhD

Laboratoire de Bioénergétique Cellulairehttp://biam.cea.fr/drf/biam/Pages/laboratoires/lbc.aspx Institut de Biosciences et Biotechnologies d’Aix-Marseille (BIAM) UMR 7265 CNRS-CEA-Université Aix-Marseille II CEA Cadarache, Bât. 156, 13108 ST PAUL LEZ DURANCE Tel: +33 (0)4 42 25 46 19 Mail [email protected]mailto:[email protected] Researchgatehttps://www.researchgate.net/profile/Caroline_Monteil

[cid:[email protected]][cid:[email protected]]

Le 14 août 2018 à 02:20, willyard <[email protected]mailto:[email protected]> a écrit :

I am trying Sequence Matrix for the first time on my Windows 7 Professional PC with 16 GB. I need to concatenate 447 gene alignments that have up to 81 samples each. The files are fairly small (under 50 KB each), but not all of them have all of the samples. I was excited to see the first few genes load into the matrix and that this useful software can handle the alignments with missing samples. But the program always hangs up with a "loading .... into memory" error. This happens on the 3rd or the 6th or the 7th sequence. I am still running the .jar file because I cannot gett the .bat file to run. I understand that I have not yet run it with the memory changed to -Xmx2000M. I think this problem running the .bat just has to do with how %windir% works to find the .jar - it has been a long time since I needed DOS. But based on the discussions in this thread, I wonder if I will have any chance assembling this size matrix. Maybe this is too big a matrix for your java-based program. Do you have any suggestions? Should I keep trying to run it with more memory? Thank you so much, Ann Willyard

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/gaurav/taxondna/issues/63#issuecomment-412710293, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AoFo40tTPkJQOpy4w98-p1a4z39KkaK6ks5uQhfJgaJpZM4JTjz7.