Thesis
Stijn van Dongen, Graph Clustering by Flow Simulation. PhD thesis, University of Utrecht, May 2000.
Download a gzipped PDF of the thesis (3.5Mb, unpacked 4.5Mb). The University of Utrecht publishes the thesis as well. They host a PDF of each separate chapter, plus the whole shebang in one piece as well. The PS file is unfortunately only useful if you have Lucida fonts installed on your system. Few people have them, so I withhold the PS source - mail me if you are interested. The thesis was written in LaTeX, and I did the uttermost to avoid that dreary default LaTeX look. This included using a different font set - it is one of the (few) shortcomings of (La)TeX that there aren't that many fonts around to choose from. But otherwise (La)TeX did a great job, and the PDF source seems to be pretty small for a 174 page thesis [is it? I am not so sure anymore]. A second reason for this modest size is that nearly all figures came from home-brewn PostScript. Some remarks on this are made on my PostScript fragment.
Reports
The meat of the thesis [constists of / was cut into] three parts which can be found at CWI's report repository. The references and CWI links are:
Stijn van Dongen. A cluster algorithm for graphs. Technical Report INS-R0010, National Research Institute for Mathematics and Computer Science in the Netherlands, Amsterdam, May 2000. Download.
Stijn van Dongen. A stochastic uncoupling process for graphs. Technical Report INS-R0011, National Research Institute for Mathematics and Computer Science in the Netherlands, Amsterdam, May 2000. Download.
Stijn van Dongen. Performance criteria for graph clustering and Markov cluster experiments. Technical Report INS-R0012, National Research Institute for Mathematics and Computer Science in the Netherlands, Amsterdam, May 2000. Download.
Local copies are A cluster algorithm for graphs, A stochastic uncoupling process for graphs, and Performance criteria for graph clustering and Markov cluster experiments.
Articles
The MCL algorithm has been successfully applied in the field of protein family detection. I co-authored a paper with Anton Enright and Christos Ouzounis. It was published as Enright A.J., Van Dongen S., Ouzounis C.A. An efficient algorithm for large-scale detection of protein families. Nucleic Acids Research 30(7):1575-1584 (2002).
The first technical report above was reworked as an article. After a strenuous and intermittently very dormant submission process started in 2000 it was belatedly published as Graph clustering via a discrete uncoupling process. Siam Journal on Matrix Analysis and Applications 30-1, p121-141, 2008.