EMMAWiki/FAQs: Difference between revisions
imported>MichaelDondrup No edit summary |
m (18 revisions) |
||
(11 intermediate revisions by one other user not shown) | |||
Line 6: | Line 6: | ||
== Giving feedback == | == Giving feedback == | ||
# '''I've found an error in your softwarre, what should I do?''' <br> First of all, make sure you <code><nowiki> RTFM </nowiki></code>. | # '''I've found an error in your softwarre, what should I do?''' <br> First of all, make sure you <code><nowiki> RTFM </nowiki></code>. If possible, use the standard bug-report form under '''Help->Submit a bug, request a feature'''. This ensures, we notice the bug ASAP, and you'll get feedback on our progress in fixing the bug. If that is not possible, send an email to the >:> emma-devel >:> mailing list. | ||
If possible, use the standard bug-report form under '''Help->Submit a bug, request a feature'''. This ensures, we notice the bug ASAP, and you'll get feedback on our progress in fixing the bug. If that is not possible, send an email to the >:> emma-devel >:> mailing list. | # '''I've found an error and I've got your number. Should I call you?''' <br> In general, don't call us, we'll call you. Please use the standard way of contacting us by [[BugZilla]] or the >:> emma-devel >:> mailing list. | ||
# '''I've found an error and I've got your number. Should I call you?''' <br> In general, don't call us | # '''Somebody told me to <code><nowiki> RTFM </nowiki></code>. What does that mean?''' <br> That's considered the geek lingo acronym for '''Read the Fine Manual'''. | ||
== Browsers == | == Browsers == | ||
Line 62: | Line 62: | ||
== Jobs and Pipelines == | == Jobs and Pipelines == | ||
<span id="pipelines"></span> | |||
# '''I have submitted an analysis job, do I have to stay and wait for it?''' <br> No, you don't. In fact, the job will run fine without any further intervention from your side. You can continue with other work in EMMA, submit more jobs or log off the computer. If you come back later, you will find all jobs in your experiment. The results are listed under ''View Results'' in the ''Basic Actions'' menue of your Experiment. | # '''I have submitted an analysis job, do I have to stay and wait for it?''' <br> No, you don't. In fact, the job will run fine without any further intervention from your side. You can continue with other work in EMMA, submit more jobs or log off the computer. If you come back later, you will find all jobs in your experiment. The results are listed under ''View Results'' in the ''Basic Actions'' menue of your Experiment. | ||
# '''My job has failed, what can I do?''' <br> First: please read the message, the job has given. Often, the message explains the reasons for a failed job. If it looks somewhat cryptic to you, it could well be a software error. In this case, and only in this case, '''don't delete the job and give feedback'''. Send an email to the >:> emma-devel >:> mailing list, giving the following information: | # '''My job has failed, what can I do?''' <br> First: please read the message, the job has given. Often, the message explains the reasons for a failed job. If it looks somewhat cryptic to you, it could well be a software error. In this case, and only in this case, '''don't delete the job and give feedback'''. Send an email to the >:> emma-devel >:> mailing list, giving the following information: | ||
Line 70: | Line 70: | ||
# '''My job stays in the ''submitted'' state for hours!''' <br> Sometimes, this can be due to a high load on the compute cluster. If you encouter this situation, '''don't resubmit the job and don't delete the job'''. It's likely that a newly submitted job will not work, either. Instead, proceed as above. | # '''My job stays in the ''submitted'' state for hours!''' <br> Sometimes, this can be due to a high load on the compute cluster. If you encouter this situation, '''don't resubmit the job and don't delete the job'''. It's likely that a newly submitted job will not work, either. Instead, proceed as above. | ||
# '''My job stays in the ''running'' state for a long time!''' <br> Often, that's just a long running job. If you suspect, it's broken anyway, don't delete the job or resubmit the job. Send an email to >:> emma-devel >:> giving all the details of the job. | # '''My job stays in the ''running'' state for a long time!''' <br> Often, that's just a long running job. If you suspect, it's broken anyway, don't delete the job or resubmit the job. Send an email to >:> emma-devel >:> giving all the details of the job. | ||
== Analysis Functions == | |||
<span id="analysisFunctions"></span> | |||
# '''How do I interpret the results of a [[NormalizationPreview]]?''' <br> A normalization preview will compute all available normalization functions on the data-sets of all arrays and provide a number of M/A-plots. M/A-plots plot the logarithmic ratio of both channel intensities (M) over the total intensity of the channels (A). M = log<sub>2</sub>(CH1)/log<sub>2</sub>(CH2) and A = 0.5*log<sub>2</sub>(CH1 + CH2). You will get a plot for each method containing all array plotted using different colors for each array. | |||
# '''And what do all these colorful lines and dots mean in the [[NormalizationPreview]]?''' <br> Have a look at this page: [[NormalizationPreview]] | |||
# '''I made a ''one-sample'' t-test. Why are there only ''NaN'' values for the ''second-group''?''' <br> '''NaN''' means ''Not a Number''. this indicates that the value is either missing, infinite or that its calculation is not appropriate. Here, the latter is the case. You made a ''one-sample'' test (having the ''two-sample'' option in the test configuration unchecked). Means: you calculate stats for a single group. As there is no second group, all values are set ''NaN''. This is because both one- and two-sample tests use the same column header ([[QuantitationTypes]]) for the resulting data-sets. For more info on the Datset Browser and data display see [[EMMAWiki/WebDocumentation/HowTos/DatasetBrowserHowTo|the DatasetBrowser HowTo]] |
Latest revision as of 07:13, 26 October 2011
FAQs
If you experience a problem or have questions about EMMA 2 please read this before sending an email
Giving feedback
- I've found an error in your softwarre, what should I do?
First of all, make sure youRTFM
. If possible, use the standard bug-report form under Help->Submit a bug, request a feature. This ensures, we notice the bug ASAP, and you'll get feedback on our progress in fixing the bug. If that is not possible, send an email to the >:> emma-devel >:> mailing list. - I've found an error and I've got your number. Should I call you?
In general, don't call us, we'll call you. Please use the standard way of contacting us by BugZilla or the >:> emma-devel >:> mailing list. - Somebody told me to
RTFM
. What does that mean?
That's considered the geek lingo acronym for Read the Fine Manual.
Browsers
- Can I use my favourite browser ?
Yes, as long as your favourite browser is one of the following:
- Mozilla 1.6 and higher
- Netscape 7 and higher
- Firefox
- Galeon
- Konqueror
All of these are available free of charge for many operating systems. We recommend to use the Firefox browser.
- Why can't I use my favourite web-browser XYZ (eg. IE from M$)?
We currently do not support all existing browsers. EMMA does not function with or looks ugly with at least:
- Netscape 4 and less
- MS IE
- lynx ;)
Trying to support more different browsers while providing a high level of interactivity requires much more effort for testing and web/javascript development. Although there are some 'standards' each vendor has its slightly different interpretation of what this 'standard' means. We decided to spare us some time which can be better invested into new features and interactivity instead of supporting any of those 'standard' browsers.
If you want to use EMMA 2 please feel free to download and install one of the suported browsers.
Web Interface
- I added Arrays to my experiment, how can I add more, the link is gone?
For a completely empty experiment, EMMA will show the following links in the experiment overview:
No arrays used in this experiment.
You can import from ArrayLims or select Arrays existing in the current project.
After importing your first arrays, you will see this link instead:
Link arrays
Of course, you may still import as many arrays as you wish. On the left you will find the following link under Basic Actions:
https://www.cebitec.uni-bielefeld.de/groups/brf/software/emma/images/import.png Import from ArrayLIMS
Jobs and Pipelines
- I have submitted an analysis job, do I have to stay and wait for it?
No, you don't. In fact, the job will run fine without any further intervention from your side. You can continue with other work in EMMA, submit more jobs or log off the computer. If you come back later, you will find all jobs in your experiment. The results are listed under View Results in the Basic Actions menue of your Experiment. - My job has failed, what can I do?
First: please read the message, the job has given. Often, the message explains the reasons for a failed job. If it looks somewhat cryptic to you, it could well be a software error. In this case, and only in this case, don't delete the job and give feedback. Send an email to the >:> emma-devel >:> mailing list, giving the following information:- the EMMA project you are using
- the experiment you run the job on
- copy/paste the data from the view-results page for that job
- My job stays in the submitted state for hours!
Sometimes, this can be due to a high load on the compute cluster. If you encouter this situation, don't resubmit the job and don't delete the job. It's likely that a newly submitted job will not work, either. Instead, proceed as above. - My job stays in the running state for a long time!
Often, that's just a long running job. If you suspect, it's broken anyway, don't delete the job or resubmit the job. Send an email to >:> emma-devel >:> giving all the details of the job.
Analysis Functions
- How do I interpret the results of a NormalizationPreview?
A normalization preview will compute all available normalization functions on the data-sets of all arrays and provide a number of M/A-plots. M/A-plots plot the logarithmic ratio of both channel intensities (M) over the total intensity of the channels (A). M = log2(CH1)/log2(CH2) and A = 0.5*log2(CH1 + CH2). You will get a plot for each method containing all array plotted using different colors for each array. - And what do all these colorful lines and dots mean in the NormalizationPreview?
Have a look at this page: NormalizationPreview - I made a one-sample t-test. Why are there only NaN values for the second-group?
NaN means Not a Number. this indicates that the value is either missing, infinite or that its calculation is not appropriate. Here, the latter is the case. You made a one-sample test (having the two-sample option in the test configuration unchecked). Means: you calculate stats for a single group. As there is no second group, all values are set NaN. This is because both one- and two-sample tests use the same column header (QuantitationTypes) for the resulting data-sets. For more info on the Datset Browser and data display see the DatasetBrowser HowTo