Dear Sir,

I am a newbie for using this great software. I am installed UGENE 1.10.1 into my CENTos box. Also, i have installed TraceTuner as external tools to help me on output fasta and qual files.

• 
  http://www.mirezone.com/ugene_1.png

If i running TraceTuner manually on terminal it was successfully output some data.

• 
  http://www.mirezone.com/ugene_2.png

So, now i am trying to apply into UGENE CMD Tools:
First, i create custom elements with CMD Tools and Write to FASTA.

• 
  http://www.mirezone.com/ugene_3.png
  http://www.mirezone.com/ugene_4.png

Next, i run the schema and get an error below:
"Unknown type of abi2fasta element"

• 
  http://www.mirezone.com/ugene_5.png

Question:
1. Can UGENE Execute any tools from outside of UGENE?
2. Any method to solve above issue?

Thanks in advanced.  ;)

Hi!

Could you please attach here or send us via email fofn.txt file, so we can reproduce the issue?

Dear Sir,

Attached file is some files for your reference:
fofn.txt
ITS_GA00285.1.a1.ab1 (sample)
ITS_GA00284.1.a1.ab1 (sample)

Thanks.  :)

---

ITS_GA00285.1.a1.ab1

---

PEH wrote on Jan 18^th, 2012 at 8:55am:

Dear Sir, Attached file is some files for your reference: fofn.txt ITS_GA00285.1.a1.ab1 (sample) ITS_GA00284.1.a1.ab1 (sample) Thanks.  :)

Thanks for your feedback. We have accepted your request and the corresponded issue has been created at our tracker. The problem will be resolved as soon as possible.

You can watch the progress at our tracker. The name of the issue is https://ugene.unipro.ru/tracker/browse/UGENE-780.

PEH wrote on Jan 18^th, 2012 at 8:55am:

Dear Sir, Attached file is some files for your reference: fofn.txt ITS_GA00285.1.a1.ab1 (sample) ITS_GA00284.1.a1.ab1 (sample) Thanks.  :)

Dear user. I have researched the problem and found the reasons of it. We have refactored recently the UGENE infrastructure for achieving a good performance, but this part of the system (workers with external tools) was not modified good. So, we will fix this problem and you could use this schema with UGENE 1.10.2 in couple of weeks.

But, I don't know if your work consists only of this schema? If it is true, then you don't need to use some external tool workers. Use can convert the format of files using standard tools of UGENE. Just make a schema: "Read sequence" -> "Write sequence".

The "Read sequence" worker can get a list of files, but not only one file. And if you put "ITS_GA00285.1.a1.ab1" and "ITS_GA00284.1.a1.ab1" to the "Read sequence" then you can convert them well.

---

PEH wrote on Jan 18^th, 2012 at 8:55am:

Dear Sir, Attached file is some files for your reference: fofn.txt ITS_GA00285.1.a1.ab1 (sample) ITS_GA00284.1.a1.ab1 (sample) Thanks.  :)

We have resolved your issue. You can try it downloading the latest snapshot on our website or wait for the 1.10.2 version.

Also I should tell you some things about your workflow.
1) The file fofn.txt contains relative paths to ABI files. So, when you use ttuner from UGENE like an external tool these paths become not correct. So, you need to write absolute paths into this file.

2) You can insert "attribute" to your worker for keeping path to fofn.txt. See screenshots.

---

Dear Sir,

I did the same way as you shown on above. But i get "Fail to create a worflow Context".

I am using:
- Ugene 1.10-0-dev (30-Jan-2012)
- Centos Linux

Attached screenshot of the error.

Thanks.  :)

---

PEH wrote on Jan 30^th, 2012 at 9:58am:

Dear Sir, I did the same way as you shown on above. But i get "Fail to create a worflow Context". I am using: - Ugene 1.10-0-dev (30-Jan-2012) - Centos Linux Attached screenshot of the error. Thanks.  :)

It very strange. Have you tried to run some simple workflow schema on your computer with this version of UGENE? For example:
Read sequence -> Write sequence
I suppose that this schema would give this error too.

The error you have received occurs if UGENE can not create the special UGENE database for keeping temporary data.
The default path to this database is "/tmp/ugene_tmp/". Maybe you should set rights on writing to this directory and everything would be all right. Try it, please.

Dear Sir,

Yes, the simple schema 'Read sequence -> Write sequence' is work.

I had giving full access for "/tmp/ugene_tmp/" but seems the access error still occur.

Attached is screenshot.  ;)

Thanks.

---

Also some problems could appear because more than one versions of UGENE are installed. These versions use one configuration file:
~/.config/Unipro/UGENE.ini
You should remove this file before you run new installed version of UGENE. It could reduce amount of problems.

PEH wrote on Jan 30^th, 2012 at 1:22pm:

Dear Sir, Yes, the simple schema 'Read sequence -> Write sequence' is work. I had giving full access for "/tmp/ugene_tmp/" but seems the access error still occur. Attached is screenshot.  ;) Thanks.

Ok, you can see that when you granted UGENE the writing access to this directory, those error had disappeared.

This message tells that the "abi2Fasta" element is working, but no data is coming to "Write sequence" and the last has no documents to write.

Have you set full paths to ABI files in fofn.txt? I told you that when you launch ttuner from UGENE then relative paths don't work. Try to write to fofn.txt absolute paths to ABI files.

Yes, i did set the ABI full path in fofn.txt. But it still happen "Error opening URL for read".

Sorry for disturbing but hope that you can help me on this issue.  ;)

---

PEH wrote on Jan 30^th, 2012 at 2:36pm:

Yes, i did set the ABI full path in fofn.txt. But it still happen "Error opening URL for read". Sorry for disturbing but hope that you can help me on this issue.  ;)

The log message is "Error opening URL for read: " and the temp file. So, after ttuner has finished its work, no files were created. Maybe ttuner has no writing access to "/tmp/ugene_tmp/" too, hasn't it?

Since i had granted full right to "/tmp/ugene_tmp/". It should not an issue right? Hmmm... something weird happen but i am not sure where it's.  :(

---

PEH wrote on Jan 30^th, 2012 at 4:43pm:

Since i had granted full right to "/tmp/ugene_tmp/". It should not an issue right? Hmmm... something weird happen but i am not sure where it's.  :(

Oh! I found! :)

Look at your fofn.txt file. It contains:
"/home/kenson/Download/ABI/forward/...."

But your folder is named "Downloads", not "Download".
Change it and try again :)

Dear Sir,

Ops, thanks for correct me. :)
Erm... After try it again and its still giving the same error saying "Error opening URL for read:"  :(

PEH wrote on Jan 31^st, 2012 at 10:43am:

Dear Sir, Ops, thanks for correct me. :) Erm... After try it again and its still giving the same error saying "Error opening URL for read:"  :(

Have you tried just to run:
./ttuner -if /home/kenson/Downloads/ABI/forward/fofn.txt -if /tmp/ugene_tmp/tmp_file.fa
?

If tmp_file.fa will be not created then it is some problem of environment, but if it will be then it is our work and we will work with it...

I have tried this task on Ubuntu with the last snaphot of UGENE (that you have too).

How have I created the new element? You can see it in attachments.
I haven't set environment variable for "ttuner", so I have typed the full path to "ttuner".
Maybe you should do it too?

---

And another part of creating of the new element.

---

Then I created a schema and run it.

---

Hmm... i got it run nicely now.

What i did is I just remove fof URL parameter and hardcode fof file path in Execution String:
ttuner -if /home/kenson/Downloads/ABI/forward/fofn.txt -sa $outfile

instead of:
ttuner -if $fof -sa $outfile

It's Ugene has issue on URL parameter?

PEH wrote on Jan 31^st, 2012 at 3:23pm:

Hmm... i got it run nicely now. What i did is I just remove fof URL parameter and hardcode fof file path in Execution String: ttuner -if /home/kenson/Downloads/ABI/forward/fofn.txt -sa $outfile instead of: ttuner -if $fof -sa $outfile It's Ugene has issue on URL parameter?

It is very good that we finally found a decision! Congratulations! :)

Sorry, but I didn't understand clear your last question.

Yes, UGENE provides setting up parameters when you are creating a cmdline element. I have already wrote about it:
http://ugene.unipro.ru/forum/YaBB.pl?num=1326766085/0#6

Maybe those problem was because you set incorrect URL to the created "fof" parameter?

By the way, this mechanism (creating external CMDLine workers) provides running some external programs from UGENE and we have no any way for controlling and retrieving errors from these programs.

I think, TraceTuner prints something into stdout about its problems when you launch your broken example, but we can not catch these messages. UGENE just runs ttuner and waits while the last one is finished up.

Thanks for your feedback. It is very necessary for us to get error reports from users :)

Sorry that i that i didn't not explain clearly on my last question. Actually i had created $fof as URL in Parameter and in execution string i enter: ttuner -if $fof -sa $outfile then i get error.

P/S: $fof browse and selected fofn.txt

After that i just remove $fof parameter and i enter execution string:
ttuner -if /home/kenson/Downloads/ABI/forward/fofn.txt -sa $outfile

Just want to know why $fof doesn't retrieve the correct path.

---------

I will check more info about ttuner whether there is log created or not.

Between, can you try to use ubuntu run below schema whether ok or not:
1. Read Sequence - select ABI file.
2. abi2fasta - as screen attached
3. Write Sequence

For above schema i got "Error opening URL for read". Just want to know why.  ;)

Thanks for helping me.  :)

---

PEH wrote on Jan 31^st, 2012 at 5:54pm:

Sorry that i that i didn't not explain clearly on my last question. Actually i had created $fof as URL in Parameter and in execution string i enter: ttuner -if $fof -sa $outfile then i get error. P/S: $fof browse and selected fofn.txt After that i just remove $fof parameter and i enter execution string: ttuner -if /home/kenson/Downloads/ABI/forward/fofn.txt -sa $outfile Just want to know why $fof doesn't retrieve the correct path. --------- I will check more info about ttuner whether there is log created or not. Between, can you try to use ubuntu run below schema whether ok or not: 1. Read Sequence - select ABI file. 2. abi2fasta - as screen attached 3. Write Sequence For above schema i got "Error opening URL for read". Just want to know why.  ;) Thanks for helping me.  :)

This schema is not so good.
First of all, ABI files are not just for keeping sequences. They keep sequences and chromatograms. As you can see from the Workflow Designer manual, UGENE Wokrflow Designer has 4 datatypes: Sequence, Annotation, MSA, String.
So, when you read an ABI file by the "Read sequence" element, its output port has no any chromatogram, it contains just a sequence from the file.

You set the "infile" parameter format as "FASTA". So, what is happening in UGENE?
1) The "Read sequence" element reads a sequence from your ABI files and sends it to "abi2fasta". The "infile" parameter format is FASTA, so the sequence is saved into tmp file with FASTA format.
2) abi2fasta runs ttuner:
./ttuner created_tmp_file.fa -sa another_tmp_file.fa
because of the "outfile" parameter format is also FASTA.

So, you can see that "./ttuner created_tmp_file.fa -sa another_tmp_file.fa" is not good execution string. You can run simple ttuner with such way and you would see that it will be not working.

-----------------

You should understand, that when you create input/output port with Sequence type and with name "qqqqq", and if you write execution string:
"<your_tool> $qqqqq"
then the real execution string will be:
"<your_tool> temporary_file.<chosen_format>"

Dear PEH,
have you got to solve your problem with adding a parameter?
http://ugene.unipro.ru/forum/YaBB.pl?num=1326766085/15#21

If you haven't, could you send us, please, a log of your failed UGENE run? (please, set all of log levels: right click at the log area -> Settings -> check all log levels). We would try to find a solution.

Our team has reviewed the mechanism of external tool elements using in UGENE and decided to improve it at several things.

You can see them through these links:
https://ugene.unipro.ru/tracker/browse/UGENE-807
https://ugene.unipro.ru/tracker/browse/UGENE-812
https://ugene.unipro.ru/tracker/browse/UGENE-815
https://ugene.unipro.ru/tracker/browse/UGENE-816
https://ugene.unipro.ru/tracker/browse/UGENE-817
https://ugene.unipro.ru/tracker/browse/UGENE-819

Dear Sir,

I'm facing the same stated problem regarding the integration of ttuner into the WD using CMD line elements. I'm using the latest UGENE 1.10.2 and its still not working. Kindly advice.

Regards,
Kenneth

kenkl wrote on Feb 10^th, 2012 at 10:47am:

Dear Sir, I'm facing the same stated problem regarding the integration of ttuner into the WD using CMD line elements. I'm using the latest UGENE 1.10.2 and its still not working. Kindly advice. Regards, Kenneth

Dear Kenneth,

could you describe your problems, please? What things can't you do with ttuner integration?

Yes.

I could reproduce the one you had posted earlier in the post, however when I set URL as a parameter to browse for my own input data it failed with an error of "error opening URL to read" and then a statement stating "nothing to write:no document".

I had the snapshots of what I had been doing but could not upload it on my post since yesterday.

Hope you can help. =)

Regards,
Kenneth

kenkl wrote on Feb 13^th, 2012 at 2:38pm:

Yes. I could reproduce the one you had posted earlier in the post, however when I set URL as a parameter to browse for my own input data it failed with an error of "error opening URL to read" and then a statement stating "nothing to write:no document". I had the snapshots of what I had been doing but could not upload it on my post since yesterday. Hope you can help. =) Regards, Kenneth

It is very strange, because I can not reproduce this problem. I have tried it on two systems (Windows XP and Ubuntu) using the latest snaphots and it works good. Could you show me in detail, please, how do you create this element (with screenshots)?

Also, it would be good if you send a log of running process (please, set all of log levels: right click at the log area -> Settings -> check all log levels).

Then I could investigate your problem better.

Hi, this is what had happened: I'm testing on the integration of ttuner into the UGENE workflow designer:

The workflow are as the first screenshot:
Read abi sequence > ttuner1 > ttuner2 > Write sequence

ttuner1: ttuner -sa $outfile -if $infile
$infile : Sequence (GFF)
$outfile: Sequence (FASTA)

ttuner2: ttuner -qa $outfile -if $infile
$infile : Sequence (GFF)
$outfile : Sequence (FASTA)

This works well when I wrote hard code with the path to the input files, but when I change it to URL input it does not work.

Kindly advice, thanks!

Regards,
Kenneth

---

Besides, there are also some errors I mentioned earlier and hereby I attach the screenshots for your reference:

---

I can comment your problems.

1) First of all, I should tell about the second schema. As I can see, your execution string is "ttuner -sa $seq -id $in". The ttuner cmdline key "-id" means that you put a directory to the input. But the "Read sequence" element reads files, and only sequence data will be available in the output port, but not any directories.

2) I have finally understood why your and PEH's attempts for using parameters doesn't work :) We call different things by one name (parameter). As I understand, you talk about cmdline parameters ($out, $in, ...). But I talk about Workflow element's parameters (see a screenshot). I will show how to use them.

And what about your first schema. I have already told about using ABI files in Workflow.
a) So far Workflow Designer doesn't support working with chromatogram data. When "Read Sequence" element reads an ABI file, containing both sequence and chromatogram, only sequence data will available in the output port.
b) So, when you set up a cmdline element's port you can choose only 4 formats for this port (Fasta, Fastq, GFF, Genbank). And if you want to set up your custom element for getting a sequence data from an input port, you should correctly set the execution string too (for example, be sure that $in is able to be a file with the port specified format).

I know that it is not so simple and clear, but your task is complex too (you want to unite together the work of two independent applications)

3) How can I help you? Please, look at my post
http://ugene.unipro.ru/forum/YaBB.pl?num=1326766085/0#6
You can create a custom element that would have no input ports and have output sequence port $out_seq (with the fasta format). Also, it would have a URL parameter $fof (Workflow element parameter  ;)). fof would store a url for a file of files. And the execution string would be: "ttuner -if $fof -sa $out_seq".

This element gets a file of files from the parameter, launches ttuner (that creates an output fasta file) and puts sequences to the output port.

---

We are working on improving this unhandy thing. When these two issues will be resolved:
https://ugene.unipro.ru/tracker/browse/UGENE-817
https://ugene.unipro.ru/tracker/browse/UGENE-819
then you would create more handy and flexible schemes.

Hi German, thanks for your clarification but unfortunately it still does not work. Please see screen shots.

Besides, will future release of UGENE WD support chromatogram data? Kindly advice, thanks!

---

Are you sure that your $www parameter contains exactly a url to a file of files? As I can see from screenshots, your $www is "/home/kennethtan/Desktop/ABI/forward/ITS_GA00016.1.a1.ab1". And I think that it is just an ABI file :)

You should create a text file "name.txt" with the next content:
/home/kennethtan/Desktop/ABI/forward/ITS_GA00016.1.a1.ab1




And some other files you need.

Also, you always could check you custom element. Just run your execution string from command line (replacing your $parameters by real files) and see the result.
Run your string:
ttuner -if /home/kennethtan/Desktop/ABI/forward/ITS_GA00016.1.a1.ab1 -sa ken.fa
and you would see that it doesn't work.

Please, try it and tell me about the results. Good luck!

Kenneth, see the attachment, please. There is an example of the input file, that you need to put to your $www parameter.

Check, that I wrote the correct url :)

---

Hi German, thanks for pointing out, it works well now! :) Anyway, will future release of UGENE WD support chromatogram data such as abi files? Please let me know asap, thanks!

kenkl wrote on Feb 14^th, 2012 at 3:18pm:

Hi German, thanks for pointing out, it works well now! :) Anyway, will future release of UGENE WD support chromatogram data such as abi files? Please let me know asap, thanks!

Good job! :)
About chromatograms: currently UGENE has no any functionality for working with chromatograms, you can just open an ABI file and see the content of the file. So, Workflow Designer has no elements for working with chromatograms too.

In next releases UGENE will not support chromatogram datatype in Workflow.

But we have planned this feature for some future releases:
https://ugene.unipro.ru/tracker/browse/UGENE-838

We will be appreciate if you tell us what functionality is needed by biologist for working with chromatograms and that we could implement in UGENE. It is better to write at "Feature Requests" forum:
http://ugene.unipro.ru/forum/YaBB.pl?board=feature

Hi! Chromatogram data processing is important for scientific research, especially in DNA sequencing. When a genome is sequence with a automated sequencing machine the raw data are produced in abi files. Thus it would be very useful for if biologist could convert the data to a more useful format such as FASTA. When it is converted then the sequence can be used for many downstream processing. Great to hear that you guys are gonna include support for abi in future release, but could it be as soon as 1.10.3 or 1.11?

Thanks!

kenkl wrote on Feb 15^th, 2012 at 4:30pm:

Hi! Chromatogram data processing is important for scientific research, especially in DNA sequencing. When a genome is sequence with a automated sequencing machine the raw data are produced in abi files. Thus it would be very useful for if biologist could convert the data to a more useful format such as FASTA. When it is converted then the sequence can be used for many downstream processing. Great to hear that you guys are gonna include support for abi in future release, but could it be as soon as 1.10.3 or 1.11? Thanks!

You are able to convert ABI to FASTA (or other formats) now, without any additional supporting and tools. Just make a scheme in Workflow Designer:

Read Sequence -> Write Fasta

Put your ABI files to the "Input file(s)" parameter of the "Read sequence" element. And set the output filename for the "Write Fasta" element. And you will receive the data in the needed useful format.

Hi German, abi files consist 2 types of data: sequence and quality score. Both data are equally important in downstream processing as some bioinformatics tools require these 2 types of files, namely seq file and qul file, to process the data.

Besides I would want to ask isit possible to input 2 data from 2 different elements upstream into a single CMDline element without increasing the number of input port?

Thanks!  :)

kenkl wrote on Feb 16^th, 2012 at 10:43am:

Hi German, abi files consist 2 types of data: sequence and quality score. Both data are equally important in downstream processing as some bioinformatics tools require these 2 types of files, namely seq file and qul file, to process the data. Besides I would want to ask isit possible to input 2 data from 2 different elements upstream into a single CMDline element without increasing the number of input port? Thanks!  :)

For keeping quality scores you can choose the FASTQ format.
Read Sequence -> Write Sequence
And choose "fastq" in the format property of the "Write Sequence" element.

What about the second question... Currently, it is not possible. But could you describe me a purpose of it? What different 2 input data and what external tool do you want to use? It is important for me to know about it for clear understanding the problem and finding the best solution.

Ok, here are the details:

I'm trying to construct a trace processing pipeline using UGENE WD. ttuner is used to call bases as well as giving quality score for each input chromatogram data. Then will use LUCY to trim off vector sequences.

Lucy is a vector sequence trimming tool and it requires at least 4 files to work:
-Vector sequence
-Splice site
-Input quality score
-Input FASTA sequence

ttuner does not output a single FASTQ file, instead it will output 2 separate files containing the quality score and FASTA sequence. Thus thus these 2 file can serve as an input data for LUCY for vector sequence removal.

Since ttuner output 2 separate files, thus I had to create 2 separate CMDline tool element for that purpose, which has been done pretty nicely. The problem arises as I could not pass 2 input files to LUCY element (created using CMDline tool), thus came my questions of whether it is possible to have 2 input data from 2 upstream elements with only 1 input port.

You can have a look at LUCY commandline through this link:
http://www.complex.iastate.edu/download/Lucy2/gadgets/lucy.pdf

Thanks for your help.

Kenneth

---

Kenneth, thanks for your feedback.
It is very good that you tell us about needed features, because they are useful and we should implement them.

Yes, your scheme shows that several slots supporting is needed for some cmdline tools. I have created an issue at our bug tracker:
https://ugene.unipro.ru/tracker/browse/UGENE-843

It will be fixed asap.

Yes, appreciate that very much!

Besides, there is another issue too regarding ttuner. I had to hardcode path to fof.txt containing abi files in order for the elements to work, and when I change the path to input from read sequence it does not work. Thus can I request an element such as "Read ABI" to be included in 1.11 or that abi file to be supported in the input sequence. THanks!

Kenneth

kenkl wrote on Feb 16^th, 2012 at 3:34pm:

Yes, appreciate that very much! Besides, there is another issue too regarding ttuner. I had to hardcode path to fof.txt containing abi files in order for the elements to work, and when I change the path to input from read sequence it does not work. Thus can I request an element such as "Read ABI" to be included in 1.11 or that abi file to be supported in the input sequence. THanks! Kenneth

This issue could be resolved using a new element "Get files". See:
https://ugene.unipro.ru/tracker/browse/UGENE-817

You can set the input directory for this element and set include or exclude filters for filenames.
This element has only one output slot (url slot). It produces paths to files in the specified directory.

Another useful improvement is a new type for input slot of custom elements ("String value"). See:
https://ugene.unipro.ru/tracker/browse/UGENE-819

You can set up your custom "ttuner" element using a way that the input is not a sequence, but it is a string value. See the first screeshot.

And this string value would not be saved to the temporary file (like it is made for all other types). This string value would be placed directly to the execution string instead of $in. So, your ttuner execution string would be:
ttuner -sa $out $in

And your scheme would be like the scheme on the second screeshot.

These improvements will be available in the 1.11 version.

---

Great to hear that! Are these features available in snapshots? Thank you very much!

kenkl wrote on Feb 16^th, 2012 at 4:49pm:

Great to hear that! Are these features available in snapshots? Thank you very much!

Yes, they will be available in snapshots in several hours or tomorrow. Now, these snapshots are being built.

Ok, thanks! :)

Hi there! I had downloaded the latest snapshot for Linux Distribution installed as 1.11-dev. There are several flaws that I had noted, error prompt in log when launched in 1.11-dev, but the other major problem is that the custom CMDline element that worked well in 1.10.2 now can't work at all in 1.11-dev. Please see screenshot. Kindly advice, thanks!

---

Besides, its good to know that in WD, workflow designed using earlier versions are not compatible to later versions. Can this issue be fixed as it would be very problematic for users that relies on custom CMDline element to re-create the element again in latest version. Thanks!

kenkl wrote on Feb 17^th, 2012 at 10:44am:

Besides, its good to know that in WD, workflow designed using earlier versions are not compatible to later versions. Can this issue be fixed as it would be very problematic for users that relies on custom CMDline element to re-create the element again in latest version. Thanks!

Workflow schemes are compatible with earlier versions.

As I can see from the screenshots your workflow doesn't work because of config file problems. Try to delete the old config file (of the 1.10.2 version):

~/.config/Unipro/UGENE.ini

Thanks! Its working well now. But its good to note that when I launch the snapshot it displayed error in log, which is not apparent in 1.10.2, see screenshot. Besides I cannot find this fixed feature in 1.11-dev:

https://ugene.unipro.ru/tracker/browse/UGENE-819

Thanks!

Kenneth

---

kenkl wrote on Feb 17^th, 2012 at 11:32am:

Thanks! Its working well now. But its good to note that when I launch the snapshot it displayed error in log, which is not apparent in 1.10.2, see screenshot. Besides I cannot find this fixed feature in 1.11-dev: https://ugene.unipro.ru/tracker/browse/UGENE-819 Thanks! Kenneth

Do you use "Linux binary package, i386 (32 bit)" or "Linux binary package, x86-64 (64 bit)"?
If you use the 32 bit version then try to download it again, please, because this version was not built yesterday but it is available now:
http://ugene.unipro.ru/downloads/unstable/ugene-1.11.0-dev-linux-i686-r1554.tar.gz

But if you use 64 bit version, this feature should be available there. Are you sure that you have chosen the "String" type for the slot? (See the screenshot).

---

Thanks Sir, no issue with that anymore as it is now working well.  :)

Now I'm trying to write my output data in either .tab / .XML format but finds that the Write Plain Text does not support tabulated format. Any idea on that? Or can come up with an element that can probably function like the "String Value" created? Output the file as it is.

Thanks!

Kenneth

Sorry for bothering again. I'm also trying to incorporate several other tools, one of such is MISA or namely Microsattelite Identification Tool. This tool uses totally different file type in its pipeline, made of 4 components which I had to separate them into components.

The problem arises as these components each receive and output different files:
misa.pl (input .fa) (output .fa.misa)
p3_in.pl (input .fa.misa) (output .p3in)
primer3 (input .fa.p3in) (output .p3out)
p3_out.pl (input .p3out) (otuput .result)

I could foresee that the WD might not be able to support such file types, could anything be done on it? Perhaps also include the "String Value" on Output slot as well?

Thanks! :)

kenkl wrote on Feb 20^th, 2012 at 9:30am:

Sorry for bothering again. I'm also trying to incorporate several other tools, one of such is MISA or namely Microsattelite Identification Tool. This tool uses totally different file type in its pipeline, made of 4 components which I had to separate them into components. The problem arises as these components each receive and output different files: misa.pl (input .fa) (output .fa.misa) p3_in.pl (input .fa.misa) (output .p3in) primer3 (input .fa.p3in) (output .p3out) p3_out.pl (input .p3out) (otuput .result) I could foresee that the WD might not be able to support such file types, could anything be done on it? Perhaps also include the "String Value" on Output slot as well? Thanks! :)

Yes, if we support the "String value" format for the output data of custom elements then users could create their elements with arbitrary input/output file formats. The corresponded issue is created:
https://ugene.unipro.ru/tracker/browse/UGENE-847
We will resolve it asap.

Thanks for your ideas, Kenneth!

No problem, thanks for such a great effort in putting UGENE together! Anyway, will this new feature be included in 1.10.3/1.11?

Thanks

Kenneth

kenkl wrote on Feb 20^th, 2012 at 4:40pm:

No problem, thanks for such a great effort in putting UGENE together! Anyway, will this new feature be included in 1.10.3/1.11? Thanks Kenneth

Yes, we will try to implement them for the 1.11 version.

Hi!

May I know how to use the "Multiplexer" element in the snapshot?

Thnaks!

Hi, I'm facing problem with one of my custom CMDline tool.

I had written the execution string (see screenshot) but it appears differently when the element is created (see screenshot). It gave an empty output file, suspect that may be due to the absent of '<' sign in the element created.

On top of that I had tried the program in commandline and it works well.

Kindly advice. Thanks!

Kenneth

---

kenkl wrote on Mar 2^nd, 2012 at 8:22am:

Hi! May I know how to use the "Multiplexer" element in the snapshot? Thnaks!

Hello!
The multiplexer element is needed to join two separate data flows into the single data flow. But a user should set a rule of data multiplexing.

There are 4 multiplexing rules:
1) 1 to 1 - for each incoming message from one input it waits for only one incoming message from another input and puts the united message to the output.

2) 1 to many - for each incoming message from the first input it waits for only every incoming message from the second input and puts the united messages to the output.

3) Many to 1 - for each incoming message from the second input it waits for only every incoming message from the second input and puts the united messages to the output.

4) Streaming mode - multiplexer transports messages as they are from inputs to the output without joining them.

There is also another parameter "If empty input". Sometimes one of the inputs (working input) receives some data but another input (closed input) becomes closed without providing any data. So, a user should set the behavior of the multiplexer for these situations. There are two variants:
1) Fill by empty values - multiplexer transports every incoming message from the working input to the output despite that the closed input receives nothing. In this case, if some of next elements should receive some data from the channel through the closed input then this element would receive empty values.

2) Truncate - don't put anything to the output if some of inputs becomes closed.

There is also a new feature which is connected with multiplexing.
In some schemes you need to divide the data flow and then to join it back using the multiplexer (see the screenshot).

The separated data could go through the elements which brake the data flow (currently, these elements are only filters).
You can see this situation on the screenshot. The data flow goes through the Filter 1 and Filter 2. So, the inputs of the multiplexer receive two different data sets, because the data flow is broken by filters.
And then these flows are joined to the single flow.

In this case, if you want to connect the sequence slot from "Read sequence" with the sequence slot of "Write sequence", you should set what path the sequence went through. As you can see from the screenshot, there are two sequences: a sequence, that goes through the Filter 1, and a sequences, that goes through the Filter 2.

---

kenkl wrote on Mar 2^nd, 2012 at 4:12pm:

Hi, I'm facing problem with one of my custom CMDline tool. I had written the execution string (see screenshot) but it appears differently when the element is created (see screenshot). It gave an empty output file, suspect that may be due to the absent of '<' sign in the element created. On top of that I had tried the program in commandline and it works well. Kindly advice. Thanks! Kenneth

Hi German, how about this issue above yea? :)

kenkl wrote on Mar 2^nd, 2012 at 4:12pm:

Hi, I'm facing problem with one of my custom CMDline tool. I had written the execution string (see screenshot) but it appears differently when the element is created (see screenshot). It gave an empty output file, suspect that may be due to the absent of '<' sign in the element created. On top of that I had tried the program in commandline and it works well. Kindly advice. Thanks! Kenneth

The problem appears because the stdin redirection ("<") is not implemented in WD. Currently, I can't answer you why ">" is implemented well, but "<" is not. Maybe there are some complex problems connected with this issue.

We will investigate it asap and tell you about the results. The corresponded issue is created:
https://ugene.unipro.ru/tracker/browse/UGENE-867

Hi! I'm just wondering is it possible that user could combine 2 programs in one single custom CMDline element. For instance ttuner that has 2 function, base calling and producing quality score. Instead of creating 2 elements can i incorporate it into the custom CMDline? Just curious on this.

Kenneth

kenkl wrote on Mar 12^th, 2012 at 10:55am:

Hi! I'm just wondering is it possible that user could combine 2 programs in one single custom CMDline element. For instance ttuner that has 2 function, base calling and producing quality score. Instead of creating 2 elements can i incorporate it into the custom CMDline? Just curious on this. Kenneth

Could you describe the problem in details? It is not so understandable what is "combine 2 programs in one element".
Describe, please, how do you want to combine programs? Or what should this element do?

Hi, thanks for replying. This inquiry arise due to the function of one tools: ttuner.

At present, I need to create 2 custom CMDline elements for ttuner, 1 for base calling, the other 1 for calling quality values. So instead of creating 2 element for those 2 functions, is it possible that I can put a semi-colon (;) for this purpose?

E.g:
ttuner -sa $infile $outfile ; ttuner -qa $infile $outfile

with the above in the execution string in a single custom CMDline element? Though I can do it with 2 elements but just an idea not sure whether this would work or not. Thanks!

Kenneth

kenkl wrote on Mar 13^th, 2012 at 2:44pm:

Hi, thanks for replying. This inquiry arise due to the function of one tools: ttuner. At present, I need to create 2 custom CMDline elements for ttuner, 1 for base calling, the other 1 for calling quality values. So instead of creating 2 element for those 2 functions, is it possible that I can put a semi-colon (;) for this purpose? E.g: ttuner -sa $infile $outfile ; ttuner -qa $infile $outfile with the above in the execution string in a single custom CMDline element? Though I can do it with 2 elements but just an idea not sure whether this would work or not. Thanks! Kenneth

We will think about that. Thanks for your ideas, Kenneth.

German wrote on Mar 5^th, 2012 at 12:30pm:

There is also a new feature which is connected with multiplexing. In some schemes you need to divide the data flow and then to join it back using the multiplexer (see the screenshot). The separated data could go through the elements which brake the data flow (currently, these elements are only filters). You can see this situation on the screenshot. The data flow goes through the Filter 1 and Filter 2. So, the inputs of the multiplexer receive two different data sets, because the data flow is broken by filters. And then these flows are joined to the single flow. In this case, if you want to connect the sequence slot from "Read sequence" with the sequence slot of "Write sequence", you should set what path the sequence went through. As you can see from the screenshot, there are two sequences: a sequence, that goes through the Filter 1, and a sequences, that goes through the Filter 2.

Hi German,

May I know how to use the Filter element and the Sequence Marker element? I'm trying to mark and filter out Blast results with certain identity threshold. Thanks!

All information about sequence markers and filters is available in our documentation:
http://ugene.unipro.ru/documentation/wd_manual/workflow_elements/data_flow/marker.html
http://ugene.unipro.ru/documentation/wd_manual/workflow_elements/data_flow/filter.html

You can set up a new marker group to the sequence marker. And the corresponding output text slot will be added to the sequence marker.

The data provided by this slot could be used at the filter element. You just need to bind the marker group slot with filter input slot; and set the "Filter by value(s)" parameter as you wish.

For example, if you set up the marker that values of the marker group are "good" and "bad" (and "rest" for unmarked data) then the "Filter by value(s)" parameter could be set as "good", or "bad", or "rest" (or comma separated values: "good, rest").