Difference between revisions of "User:Dennis9600"

From EteRNA WiKi
Jump to: navigation, search
Line 1: Line 1:
<p><span style="font-size: medium;">11/4/2013: I am struggling with the current round of active labs.&nbsp; Two issues are just killing my analysis tools: </span></p>
+
<p><span style="font-size: medium;"><strong>Analysis Reports and Spreadsheets produced by my Python script can be found in <a href="https://drive.google.com/folderview?id=0B-rDnoMjFSH2cDFibWt6T19nRzg&amp;usp=sharing">ReportsActiveLabs</a>.</strong></span></p>
<p><span style="font-size: medium;">1) There are non-canonical pairs in the target structures of many of the labs, e.g A-A pairs. &nbsp; </span></p>
+
<p><span style="font-size: medium;">2) The "tails" that are added to the hook aren't the same any more.&nbsp; They differ from lab to lab and I haven't figured out how to automate the correct generation of the tails.</span><span style="font-size: medium;">&nbsp; Their doesn't appear to be any specification of the tails in the JSON data returned by the html GETs that are available.</span></p>
+
<p>&nbsp;</p>
+
<p><span style="font-size: medium;">If you think you have an answer to any of the above issues, please help me out by sending me a PM.&nbsp; Thanks in advance for any help!</span></p>
+
<p>&nbsp;</p>
+
 
<p>&nbsp;</p>
 
<p>&nbsp;</p>
 +
<p><span style="font-size: medium;">11/4/2013: The results posted so far do not cover all of the active labs for this cycle.&nbsp; :-( </span></p>
 +
<p><span style="font-size: medium;">I have been struggling with the current round of active labs.&nbsp; Two issues have been killing my analysis tools: </span></p>
 +
<p><span style="font-size: medium;">1) There are non-canonical pairs in the target structures of many of the labs, e.g A-A pairs. I am still working on this issue...<br /></span></p>
 +
<p><span style="font-size: medium;">2) The "tails" that are added to the hook aren't the same any more.&nbsp; They differ from lab to lab.&nbsp; I have a fix for this issue and you will start to see more and more lab results posted in the folder mentioned above.&nbsp; Thanks to ElNando for info that lead to the fix!<br /></span></p>
 
<p><span style="font-size: medium;"><br /></span></p>
 
<p><span style="font-size: medium;"><br /></span></p>
 +
<p><span style="font-size: medium;">If you think you have an answer/suggestion for issue #1 above, please help me out by sending me a PM.&nbsp; Thanks in advance for any help!</span></p>
 
<p>&nbsp;</p>
 
<p>&nbsp;</p>
<p><span style="font-size: small;">The links below take you to analysis by my lab tool on the currently active labs.</span></p>
+
<p><span style="font-size: small;"><strong>In this edition:</strong></span></p>
 +
<p><span style="font-size: small;"><strong>&nbsp;</strong>1)&nbsp; there is a field for the Vienna 2.1.1 melt point of each design submitted.&nbsp; This new field now appears in both the text reports and the spreadsheets.<br /></span></p>
 +
<p><span style="font-size: small;">My lab tool includes a forecasting tool that looks at factors that have correlated well with past synthesis scores.&nbsp; The current version of the forecasting tool looks at the following factors:</span></p>
 +
<p><span style="font-size: small;">a) Whether or not the design folded correctly in EteRNA's energy model (Vienna 1.8.5)</span></p>
 +
<p><span style="font-size: small;">b) Whether or not the design folded correctly in the Vienna 2.1.1 Energy Model</span></p>
 +
<p><span style="font-size: small;">c) The melt point of the design (as reported by the EteRNA server)</span></p>
 +
<p><span style="font-size: small;">d) The percentages of C,U, and G in the design, and how far they differ from 13, 10, and 21% respectively (the Berex Strategy).</span></p>
 +
<p><span style="font-size: small;">e) log<sub>10</sub>(designer's EteRNA points).&nbsp; This factor is given a very low weight.<br /></span></p>
 
<p>&nbsp;</p>
 
<p>&nbsp;</p>
<p><span style="font-size: large;"><strong>The most current set of reports and spreadsheets posted October 22 is in&nbsp;<a href="https://drive.google.com/folderview?id=0B-rDnoMjFSH2VkRKYWZrSGJRMEE&amp;usp=sharing">ReportsActive2013Oct22</a></strong></span></p>
+
<p><span style="font-size: small;"><strong>Things I have looked at:</strong><br /></span></p>
 +
<p><span style="font-size: small;">I have also looked at a) temperature and b) designer's EteRNA points as possible factors for my forecasting tool.&nbsp; Temperature setting of the Energy Model correlated best with past synthesis scores when it was set to the default of 37C.&nbsp; Designer's EteRNA points had a very weak positive correlation with past synthesis scores so is not used in the forecasting tool.</span></p>
 
<p>&nbsp;</p>
 
<p>&nbsp;</p>
<p><span style="font-size: small;">In this edition there is a field for the Vienna 2.1.1 melt point of each design submitted.&nbsp; This new field made it into the text reports, but not the spreadsheets.&nbsp; (Something to do for the next publication. &nbsp;Or the next one after that <img title="Smile" src="/wiki/extensions/TinyMCE_MW/jscripts/tiny_mce/plugins/emotions/img/smiley-smile.gif" border="0" alt="Smile" />)<br /></span></p>
 
<p><span style="font-size: small;">My lab tool includes a forecasting tool that looks at factors that have correlated well with past synthesis scores.&nbsp; The current version of the forecasting tool looks at the following factors:</span></p>
 
<p><span style="font-size: small;">1) Whether or not the design folded correctly in EteRNA's energy model (Vienna 1.8.5)</span></p>
 
<p><span style="font-size: small;">2) Whether or not the design folded correctly in the Vienna 2.1.1 Energy Model</span></p>
 
<p><span style="font-size: small;">3) The melt point of the design (as reported by the EteRNA server)</span></p>
 
<p><span style="font-size: small;">4) The percentages of C,U, and G in the design, and how far they differ from 13, 10, and 21% respectively (the Berex Strategy).</span></p>
 
<p><span style="font-size: small;">I have also looked at a) temperature and b) designer's EteRNA points.&nbsp; Temperature setting of the Energy Model correlated best with past synthesis scores when it was set to the default of 37C.&nbsp; Designer's EteRNA points had a very weak positive correlation with past synthesis scores so is not used in the forecasting tool.</span></p>
 
 
<p>&nbsp;</p>
 
<p>&nbsp;</p>
 +
<p><strong><span style="font-size: medium;">Things to do:</span></strong></p>
 +
<p><strong><span style="font-size: medium;">I am currently working on a change to my scripts that will do a better job of analyzing labs that have non-canonical pairs locked into the target structure.&nbsp;&nbsp; The Vienna RNA folding tools have an option to allow non-standard pairs.&nbsp; This option gives them a pairing free energy of 0, which may not be correct, but may be better than nothing.<br /></span></strong></p>
 
<p><span style="font-size: large;">I am (still) thinking about how to add Vienna 2.1.1 melt curves and dot plots to these reports.&nbsp; I can modify my scripts to generate these, but they are in Postscript format.&nbsp; I haven't yet figured out how to incorporate them quickly in my reports and spreadsheets.&nbsp; If you think there is any easy way to do this, please let me know.<br /></span></p>
 
<p><span style="font-size: large;">I am (still) thinking about how to add Vienna 2.1.1 melt curves and dot plots to these reports.&nbsp; I can modify my scripts to generate these, but they are in Postscript format.&nbsp; I haven't yet figured out how to incorporate them quickly in my reports and spreadsheets.&nbsp; If you think there is any easy way to do this, please let me know.<br /></span></p>
 
<p>&nbsp;</p>
 
<p>&nbsp;</p>

Revision as of 02:02, 5 November 2013

Analysis Reports and Spreadsheets produced by my Python script can be found in ReportsActiveLabs.

 

11/4/2013: The results posted so far do not cover all of the active labs for this cycle.  :-(

I have been struggling with the current round of active labs.  Two issues have been killing my analysis tools:

1) There are non-canonical pairs in the target structures of many of the labs, e.g A-A pairs. I am still working on this issue...

2) The "tails" that are added to the hook aren't the same any more.  They differ from lab to lab.  I have a fix for this issue and you will start to see more and more lab results posted in the folder mentioned above.  Thanks to ElNando for info that lead to the fix!


If you think you have an answer/suggestion for issue #1 above, please help me out by sending me a PM.  Thanks in advance for any help!

 

In this edition:

 1)  there is a field for the Vienna 2.1.1 melt point of each design submitted.  This new field now appears in both the text reports and the spreadsheets.

My lab tool includes a forecasting tool that looks at factors that have correlated well with past synthesis scores.  The current version of the forecasting tool looks at the following factors:

a) Whether or not the design folded correctly in EteRNA's energy model (Vienna 1.8.5)

b) Whether or not the design folded correctly in the Vienna 2.1.1 Energy Model

c) The melt point of the design (as reported by the EteRNA server)

d) The percentages of C,U, and G in the design, and how far they differ from 13, 10, and 21% respectively (the Berex Strategy).

e) log10(designer's EteRNA points).  This factor is given a very low weight.

 

Things I have looked at:

I have also looked at a) temperature and b) designer's EteRNA points as possible factors for my forecasting tool.  Temperature setting of the Energy Model correlated best with past synthesis scores when it was set to the default of 37C.  Designer's EteRNA points had a very weak positive correlation with past synthesis scores so is not used in the forecasting tool.

 

 

Things to do:

I am currently working on a change to my scripts that will do a better job of analyzing labs that have non-canonical pairs locked into the target structure.   The Vienna RNA folding tools have an option to allow non-standard pairs.  This option gives them a pairing free energy of 0, which may not be correct, but may be better than nothing.

I am (still) thinking about how to add Vienna 2.1.1 melt curves and dot plots to these reports.  I can modify my scripts to generate these, but they are in Postscript format.  I haven't yet figured out how to incorporate them quickly in my reports and spreadsheets.  If you think there is any easy way to do this, please let me know.

 

Please let me know if you have any other things you would like to see in my lab reports.

 

Happy folding,

---Dennis9600

Personal tools
Main page
Introduction to the Game