namics of protein distribution is a critical advantage of fluo-
In this vein, we can consider ways of representing genera-
tive models and choosing their characteristics. The models
proposed in this paper can be put into a directed probabilistic
graphical model framework, which is also known as Bayesian
network (Fig. 14). The advantage of using a graphical model is
that we can tune the model structure in a more intuitive way.
In the graphical model, each node is a component of the
model and each edge is the correlation between the nodes. The
arrow means the direction of determination. So the procedure
of model design becomes adding or removing nodes or edges.
If we consider each component as a set of random variables,
then the graph becomes a Bayesian network. Therefore we can
use well-developed techniques for Bayesian networks to do
inference and interpretation.
The goal of building the generative models is to provide
an interface between location proteomics and systems biology,
so we have begun implementing generative models in our
Protein Subcellular Localization Image Database (PSLID,
http://pslid.cbi.cmu.edu). We have also done some prelimi-
nary work to convert the models into XML format, which we
hope to merge into standard cell modeling descriptions such
as SBML (40) and CELLML (41). This will make our models
easily transferable between programs. We expect shortly to
release software to permit training of models and synthesis of
images on a variety of platforms (I. Cao-Berg, T. Zhao, and
R.F. Murphy, in preparation). We anticipate a wide applicabil-
ity of these tools in systems biology studies, especially in simu-
lations of cell behavior that require detailed models for subcel-
We thank Eric Xing and Geoffrey Gordon for helpful dis-
cussions and critical reading of this manuscript.
1. Ideker T, Galitski T, Hood L. A new approach to decoding life: Systems biology.
Annu Rev Genomics Hum Genet 2001;2:343–372.
2. Kitano H. Computational systems biology. Nature 2002;420:206–210.
3. Sauro HM, Hucka M, Finney A, Wellock C, Bolouri H, Doyle J, Kitano H.
Next generation simulation tools: the Systems Biology Workbench and BioSPICE
integration. OMICS 2003;7:355–372.
4. Faust M, Montenarh M. Subcellular localization of protein kinase CK2. A key to its
function? Cell Tissue Res 2000;301:329–340.
5. Ortoleva P, Berry E, Brun Y, Fan J, Fontus M, Hubbard K, Jaqaman K, Jary-
mowycz L, Navid A, Sayyed-Ahmad A, Shreif Z, Stanley F, Tuncay K, Weitzke E,
Wu LC. The karyote physico-chemical genomic, proteomic, metabolic cell modeling
system. OMICS 2003;7:269–283.
6. Chou K-C, Elrod DW. Protein subcellular location prediction. Prot Eng 1999;12:107–
7. Chou KC, Cai YD. Prediction and classification of protein subcellular location-
sequence-order effect and pseudo amino acid composition. J Cell Biochem 2003;
8. Park KJ, Kanehisa M. Prediction of protein subcellular locations by support vector
machines using compositions of amino acids and amino acid pairs. Bioinformatics
9. Pan YX, Zhang ZZ, Guo ZM, Feng GY, Huang ZD, He L. Application of
pseudo amino acid composition for predicting protein subcellular location: Stochas-
tic signal processing approach. J Prot Chem 2003;22:395–402.
10. Chen X, Velliste M, Murphy RF. Automated interpretation of subcellular patterns in
fluorescence microscope images for location proteomics. Cytometry Part A 2006;
11. Glory E, Murphy RF. Automated subcellular location determination and high
throughput microscopy. Developmental Cell 2007;12:7–16.
12. Boland MV, Murphy RF. A neural network classifier capable of recognizing the pat-
terns of all major subcellular structures in fluorescence microscope images of HeLa
cells. Bioinformatics 2001;17:1213–1223.
13. Murphy RF, Velliste M, Porreca G. Robust numerical features for description and
classification of subcellular location patterns in fluorescence microscope images.
J VLSI Signal Process 2003;35:311–321.
14. Velliste M, Murphy RF. Automated determination of protein subcellular locations from
3D fluorescence microscope images. In: Proceedings of the 2002 IEEE International Sym-
posium on Biomedical Imaging, Washington, DC, 7–10 June 2002. pp 867–870.
15. Chen X, Velliste M, Weinstein S, Jarvik JW, Murphy RF. Location proteomics—
Building subcellular location trees from high resolution 3D fluorescence microscope
images of randomly-tagged proteins. Proc SPIE 2003;4962:298–306.
16. Chen X, Murphy RF. Objective clustering of proteins based on subcellular location
patterns. J Biomed Biotechnol 2005;2005:87–95.
17. Hu Y, Carmona J, Murphy RF. Application of temporal texture features to automated
analysis of protein subcellular locations in time series fluorescence microscope
images. In: Proceedings of the 2006 IEEE International Symposium on Biomedical
Imaging, Arlington, VA, 6–9 April 2006. pp 1028–1031.
18. Krause A, Stoye J, Vingron M. Large scale hierarchical clustering of protein
sequences. BMC Bioinformatics 2005;6:15.
19. Balaji S, Srinivasan N. Use of a database of structural alignments and phylogenetic
trees in investigating the relationship between sequence and structural variability
among homologous proteins. Prot Eng 2001;14:219–226.
20. Loew LM, Schaff JC. The virtual cell: A software environment for computational
cell biology. Trends Biotechnol 2001;19:401–406.
21. Coggan JS, Bartol TM, Esquenazi E, Stiles JR, Lamont S, Martone ME, Berg
DK, Ellisman MH, Sejnowski TJ. Evidence for ectopic neurotransmission at a neu-
ronal synapse. Science 2005;309:446–451.
22. Huang K, Murphy RF. Boosting accuracy of automated classification of fluorescence
microscope images for location proteomics. BMC Bioinformatics 2004;5:78.
23. Thomas CH, Collier JH, Sfeir CS, Healy KE. Engineering gene expression and pro-
tein synthesis by modulation of nuclear shape. Proc Natl Acad Sci USA 2002;4:1972–
24. Blum H. Biological shape and visual science. J Theor Biol 1973;38:205–287.
25. Tam R, Heidrich W. Shape simplification based on the medial axis transform. In:
Proceedings of the 14th IEEE Conference on Visualization, 2003; Seattle, Washington,
USA. pp 481–488.
26. Hiransakolwong N, Vu K, Hua KA, Lang S-D. Shape recognition based on the
medial axis approach. Proceedings of the 2004 IEEE International Conference on
Multimedia Exposition, 2004; Taipei, Taiwan. pp 257–260.
27. Murata S-i, Herman P, Lakowicz JR. Texture analysis of fluorescence lifetime images
of AT- and GC-rich regions in nuclei. J Histochem Cytochem 2001;49:1443–1451.
28. Palcic B. Nuclear texture: Can it be used as a surrogate endpoint biomarker? J Cellu-
lar Biochem 1994;19(Suppl):40–46.
29. Jørgensen T, Yogesan K, Tveter KJ, Skjørten F, Danielsen HE. Nuclear texture anal-
ysis: A new prognostic tool in metastatic prostate cancer. Cytometry 1998; 24:277–283.
30. Zhu SC, Wu Y, Mumford D. Filters, random fields and maximum entropy (FRAME):
Towards a unified theory for texture modeling. Int J Comput Vision 1998;27:107–
31. Nealen A, Alexa M. Hybrid texture synthesis. In: Proceedings of the 14th Euro-
graphics Workshop Rendering, 2003; Leuven, Belgium. pp 97–105.
32. Portilla J, Simoncelli EP. A parametric texture model based on joint statistics of com-
plex wavelet coefficients. Int J Computer Vision 2000;40:49–71.
33. Lehmussola A, Selinummi J, Ruusuvuori P, Niemisto
, Yli-Harja O. Simulating fluo-
rescent microscope images of cell populations. In: Proceedings of the 27 Annual Con-
ference of the IEEE Engineering in Medicine and Biology Society, 2005; Shanghai,
China. pp 3153–3156.
34. Cootes TF, Taylor CJ, Cooper DH, Graham J. Active shape models—Their train-
ing and application. Comput Vision Image Understanding 1995;61:38–59.
35. Zhao T, Velliste M, Boland MV, Murphy RF. Object type recognition for automated
analysis of protein subcellular location. IEEE Trans Image Process 2005;14:1351–1359.
36. Bilmes J. A gentle tutorial on the EM algorithm and its application to parameter esti-
mation for Gaussion mixture and hidden Markov models. Berkeley, CA: Interna-
tional Computer Science Institute; 1997. Report nr TR-97–021.
37. Huang K, Velliste M, Murphy RF. Feature reduction for improved recognition of
subcellular location patterns in fluorescence microscope images. Proc SPIE
38. Garcia Osuna E, Hua J, Bateman N, Zhao T, Berget P, Murphy R. Large-scale
automated analysis of protein subcellular location patterns in randomly-tagged 3T3
cells. Ann Biomed Eng 2007;35:1081–1087.
39. Schubert W, Bonnekoh B, Pmmer AJ, Philipsen L, Bockelmann R, Malykh Y,
Gollnick H, Friedenberger M, Bode M, Dress AWM. Analyzing proteome topol-
ogy and function by automated multi-dimensional fluorescence microscopy. Nat Bio-
40. Hucka M, Finney A, Sauro H, Bolouri H, Doyle J, Kitano H, Arkin A, Bornstein B,
Bray D, Cornish-Bowden A, Cuellar AA, Dronov S, Gilles ED, Ginkel M, Gor V,
Goryanin II, Hedley WJ, Hodgman TC, Hofmeyr JH, Hunter PJ, Juty NS, Kasberger
JL, Kremling A, Kummer U, Le Novere N, Loew LM, Lucio D, Mendes P, Minch E,
Mjolsness ED, Nakayama Y, Nelson MR, Nielsen PF, Sakurada T, Schaff JC, Shapiro
BE, Shimizu TS, Spence HD, Stelling J, Takahashi K, Tomita M, Wagner J, Wang J.
The systems biology markup language (SBML): A medium for representation and
exchange of biochemical network models. Bioinformatics 2003;19:524–531.
41. Lloyd CM, Halstead MDB, Nielsen PF. CellML: Its future, present and past. Prog
Biophys Mol Biol 2004;85:433–450.
990 Generative Models for Subcellular Location