XML for Chemistry and Biosciences

Line 1 is the processing instruction letting the processor know that this is an XML file. Line 3 is the top level sytlesheet tag that encloses the whole sytle sheet. It includes a namespace attribute. The mix of XSLT tags, XML tags, and HTML tags here illustrates the importance of namespaces for differentiating the different varieties of markup. Line 5 is an output instruction letting the processor know that the output will he a HTML file and please indent it to make it easier for humans to read.

Line 8 starts an XSL template block, which is the top level template to output the outline of the HTML file. The HTML outline is given in lines 9 through 17. Line 15 tells the processor to look for other rules to apply. The next rule to be found by the XSLT processor will be the template for molecule on line 21. This outputs a h2 level 2 HTML heading tag with the text 'Molecule' and the value of the title attribute in the molecule. The block in lines 27 through 30 match an atomArray tag. They output a level 3 HTML header tag h3 and then tell the processor to apply remaining templates. The remaining template on lines 33 through 35 match an atom tag. It output the element type given in the string tag. This is not a comprehensive XSLT style sheet and many tags are ignored.

Let's use this stylesheet to transform the CML v1 file arginine.cml. The first few lines of the file are shown here.

Now we need an XSLT processor. The Sun Java 5.0 platform includes XSL processing features⁵. A number of other vendors and open source projects have XSLT processors, including Microsoft. I am using a Java XSLT processor but you may use anything other type with the XSL style sheet below. The Java file is XmlTransformer.java.

1    import java.io.File;
2    import javax.xml.transform.Result;
3    import javax.xml.transform.Transformer;
4    import javax.xml.transform.TransformerConfigurationException;
5    import javax.xml.transform.TransformerException;
6    import javax.xml.transform.TransformerFactory;
7    import javax.xml.transform.stream.StreamResult;
8    import javax.xml.transform.stream.StreamSource;
9
10   /*
11    * Creates an XSLT transformer and applies it to the first argument. The resulting xml document is
12    * output as the file output.xml.
13    */
14   public class XmlTransformer {
15       /**
16    * @param argv[0] is the name of the style sheet, argv[1] is the name of the input html file,
17    *        argv[2] is the name of the output file.
18    */
19       public static void main(String[] argv) {
20           System.out.println("Source style sheet: " + argv[0]);
21           System.out.println("Input xml document: " + argv[1]);
22           System.out.println("Output html document: " + argv[2]);
23           TransformerFactory transformerFactory = TransformerFactory.newInstance();
24       StreamSource xsltSource = new StreamSource(new File(argv[0]));
25         StreamSource xmlSource = new StreamSource(new File(argv[1]));
26         Result result = new StreamResult(new File(argv[2]));
27           try {
28              Transformer transformer = transformerFactory.newTransformer(xsltSource);
29              transformer.transform(xmlSource, result);
30          } catch (TransformerConfigurationException e) {
31             e.printStackTrace();
32     } catch (TransformerException e) {
33            e.printStackTrace();
34         }
35       }
37   }

This program loads an XSLT transformer with a given style sheet and transforms the input file to an output file. In lines 1 through 8 the relevant input files are imported. Line 14 declares the class and line 19 declares the main method. Line 23 instantiates the transformer factory, which will create the transformer. Line 24 creates the input source that the XSLT file will be read from using the first argument from the command line. Line 25 creates the input source that the input XML file will be read from using the second parameter from the command line. Line 26 defines the result that the output file will be written to using the third argument from the command line. Line 28 uses the transformer factory to create a new transformer with the XSLT input source. Line 29 takes the input XML file and transforms it to the output file. Lines 30 through 33 catch and process exceptions that may be generated during the program.

> javac src\XmlTransformer.java
> java -cp classes XmlTransformer cml1toxhtml.xslt arginine.cml arginine.html

<html>
<head>
<META http-equiv="Content-Type" content="text/html; charset=UTF-8">
<title>Extract from a Chemical Markup Language file</title>
</head>
<body>
<h1>Extract from a Chemical Markup Language File</h1>
<h2>Molecule ARGININE</h2>
<h3>Atom Array</h3>

<br>atom id: a1, C
    <br>atom id: a2, C
    <br>atom id: a3, C
    <br>atom id: a4, C
    <br>atom id: a5, C
    <br>atom id: a6, N
    <br>atom id: a7, N
    <br>atom id: a8, O
    <br>atom id: a9, O
    <br>atom id: a10, C
    <br>atom id: a11, N
    <br>atom id: a12, N
    <br>atom id: a13, H
</body>
</html>

The article Tools for Working with Chemical XML provides an XSLT stylesheet for transforming CML1 to CML2. The W3C The Extensible Stylesheet Language Family (XSL) Architecture Domain is a fundamental reference for this¹⁰. The current version is 1.0 and there is a candidate recomendation for a version 1.1. Other references are Kay, XSLT: A Programmer's Reference¹¹ and the Xalan Apache Project¹².

Other XML Languages for Chemistry and BioSciences

There are a number of of XML languages that are relevant to biosciences are listed in the XML for Molecular Biology web site¹³, including

An interesting research project is the W3C Semantic Web Health Care and Life Sciences Interest Group. Languages and projects include

Introduction to Extensible Markup Language for Chemistry and Biosciences (Continued)

Transforming XML into other formats

Other XML Languages for Chemistry and BioSciences

About the Author

Related Reading

References