Sina abruptly ends with "Killed" and no information. #74
Comments
|
Sorry to hear this. Will try to help. Might be a week or two until I can fully dive into it.
Could be out of memory. Can you check`dmesg` and/or append `; echo $?`. Also, watch with `top` from a second terminal.
If that doesn’t help, try running in gdb: `gdb --args sina ...` then enter `run` and once it crashes `bt`.
If you add `-v` you’ll see each sequence as it’s processed. Could be there is one it doesn’t like.
Elmar
…--
message sent from mobile device
On Jul 7, 2019, at 7:45 AM, Torkel ***@***.***> wrote:
So I have some files with rRNAs which I wish to align (in fasta files, different domains and type of rRNAs, but always the same in each file).
I try to use sina and the following command (on ubuntu 18.04, conda version 6.11.4, sina version 1.6.0, python version 3.7.3):
sina -i Archaea/5SrRNAs.fasta -o Archaea/5SrRNAs_aligned.fasta --db SILVA_132_LSUParc_12_12_17_opt.arb
Things then starts up nicely, but in the middle of building index it just ends with the word "Killed", and no information of why. I have tried several times, here is the results:
13:39:21 [famfinder] Option --ptdb deprecated; please use --db instead
13:39:37 [Search (internal)] No cached index found.
Processing: 0 [00:00:14]████████████████| 907382/907382 [00:00:00 / 00:00:00]
Building Index: 75% |█████████████▋ | 684833/907382 [00:22:30 / 00:07:18]Killed
14:06:01 [Search (internal)] No cached index found.
Processing: 0 [00:00:15]██████████████████████████████████████████| 907382/907382 [00:00:00 / 00:00:00]
Killedng Index: 74% |████████████████████████████████▌ | 670767/907382 [00:21:41 / 00:07:39]
(just copied two attempts out from the terminal)
However, I do not actually get any information on what have gone wrong. I would be happ if someone could help me get the program to work.
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or mute the thread.
|
|
Thanks for the quick reply (unlike my, I will blame a travel). I did run top when I did my last tests, %CPU hitting about 125 and %MEM slowly working up to ~50 before the end. At least this is what I remember, might only have been when I was looking at errors in my old sina 1.2 installation. Anyway, when I am double checking this now the error is gone, and the alignment seem to be successful. I'm confused, but it is good. I will continue to run alignments on all my files now, but hopefully things continue to turn out well. |
|
Just updating again. I have tried aligning several datasets (from Archaea, Bacteria, and Eukaryote). All types of rRNAs as well (5S, 5.8S, 16S, 18S, 23S, and 28S). To run the program I use this line (after activating sina in the terminal). [Domain] is either Arcahea, Bacteria, or Eukaryote. [Dataset Nbr] just designates one of my datasets), [rRNA type] is either 5S, 16S, or 23S (for Arachaea and Bacteria) or 5S, 5.8S, 18S, or 28S (for Eukaryote). [Database] is the reference database. I have donwloaded these databases from silva. For 16 and 18S I use "SILVA_132_SSURef_NR99_13_12_17_opt.arb" and for 5S, 5.8S, 23S, and 28S I use "SILVA_132_LSURef_07_12_17_opt.arb". When I run and watch the "top" infromation %CPU hit something like 750 and %MEM peaks at 65, but is usually around 40. If I run (That's the last few lines of a rather large output) If I try using gdb (on the third datasets that fails, some bacterial 23S rRNAs) I get this outpu: Next, when I run When I run the same dataset, but adding the Finally I show the output after the fails from the various files that failed. In each case, if I rerun the program they all fail at a similar place place (I show the output of a second run, hence the dual texts for each dataset). A set of bacterial 23S rRNAs. A set of bacterial 16S rRNAs. A set of bacterial 23S rRNAs. A set of eukaryote 28S rRNAs. A set of eukaryote 5.8S rRNAs. (Progress bars probably of uneven size due to different sizes of the terminal winodw when I copied them out) I hope this can be of some help, if you want I could provide files as well. |
|
Hi @TorkelE, sorry for the very late response. The For the large SILVA SSU reference databases, SINA just needs a lot of memory to fit the database. If I ever find the time, I might write something to use an on-disk database, but that'll be slower. |
|
Thank you very much. |
|
How much memory do you have? You might be able to either use the LTP dataset, or you could try downsampling the reference (something like |
|
At that time I only had about 16GB RAM I believe. I've recently upgraded to a new computer with 128GB Ram (but haven't tried with that). Maybe that should be enough? |
|
That's definitely ample. The 16GB should actually have been enough - my laptop has no more and is working fine for most things. |
|
Closing this in favor of #87 |
So I have some files with rRNAs which I wish to align (in fasta files, different domains and type of rRNAs, but always the same in each file).
I try to use sina and the following command (on ubuntu 18.04, conda version 6.11.4, sina version 1.6.0, python version 3.7.3):
Things then starts up nicely, but in the middle of building index it just ends with the word "Killed", and no information of why. I have tried several times, here is the results:
(just copied two attempts out from the terminal)
However, I do not actually get any information on what have gone wrong. I would be happ if someone could help me get the program to work.
The text was updated successfully, but these errors were encountered: