SOP's and the BinBase Output

 
the sop's are basically the heart ouf our export system. They define what happens with the data and how are data automatically cleaned and sized down. For example they remove not relevant data from your result file.

what are sops 

SOP's are standard operation procedures and are the input configurations for our export module. They are used to convert the native xml format into file formats which are easily to understand for humans. Like xls or txt. They also provide the possibility to remove some data in form of different filter. For example you want to split your result into two results. One result file which contains only internal standards and one result file which contains all the other data.

In other words the export does not work without a minmal sop which tells us to take the result and put it into a zip archive which can be accessed afterwards.

the default sop

This is the BinBase default sop which does the following

• remove bins which are not found to 80% in at least one class
• create an xls and txt file with the data
• defines that we are interested in the attribute height
• defines that we want to group bins which are the same compound, but were wrongly annotated because of unlucky settings

and thats how it looks

The header has to be defined like this and should not be changed since it's hard coded in the algorithm. This is a known programming error and we hope to fix it in a future version. 

<?xml version="1.0" encoding="UTF-8"?>
<sop desc="transformation of samples" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
	<!-- transform to crosstable -->
	<transform sizedown="80" attribute="height" combine="true">
		<!--the needed headers -->
		<header>
			<param value="retention_index"/>
			<param value="quantmass"/>
			<param value="id"/>
			<param value="spectra"/>
		</header>

		<rawdata folder="rawdata">
			<format type="XLS"/>
			<format type="TXT"/>
		</rawdata>
	</transform>

</sop>

I you want to modify this sop, you can find it on the application server in the configured directory. The related jmx-bean is binbase:export. This jmx gives you the possibility to upload another sop or replace the default sop. Default means this sop is always used if you don't set the sop attribute in the ExperimentClass.

If you want to know how to modify and sop please have a look at it's schema. 

How does an sop look with all the currently available features? This is an example which documents the best what it can do.

 TODOP

Why should you write your own SOP and not just use the default sop.

 The default SOP is great for exporting your data, but sometimes you need to have other reports generated and want to pre process your data for your statistical analysis like

• replace the empty values with real values based on our third pass annotation algorithm
•  generate a list of the dates when the sample were measured
• pre cache all the bins for web based queries in binbase compound browser
• have different downsizing settings like 10% and 90%
• export the retention times of the annotated peaks
• calculate similarity matrixes