Tracking T and B cells from two-photon microscopy imaging using constrained SMC clusters D. Olivieri 1* , J. Faro 2,3 , I. Gomez-Conde 1 , C. E. Tadokoro 3 1 University of Vigo, ESEI, 32004 Ourense, Spain, http://www.uvigo.es 2 University of Vigo, Department of Biochemistry, Genetics and Immunology, 36310 Vigo, Spain, http://www.uvigo.es 3 Instituto Gulbenkian de Ciencia, 2781-901 Oeiras, Portugal, http://www.igc.gulbenkian.pt Summary This paper describes a novel software algorithm, called constrained Sequential Monte Carlo (SMC) clusters, for tracking a large collection of individual cells from intra-vital two-photon microscopy image sequences. We show how our method and software tool, implemented in python, is useful for quantifying the motility of T and B lymphocytes in- volved in an immune response vs lymphocytes under non immune conditions. We describe the theory behind our algorithm and briefly discuss the architecture of our software. Fi- nally, we demonstrate both the functionality and utility of software by applying it to two practical examples from videos displaying lymphocyte motility in B cell zones (follicles) and T cell zones of lymph nodes. 1 Introduction Recent advances in intra-vital multi-photon laser microscopy have made it possible to visualize cellular motion within lymphoid tissues in real time and in living animals [1, 2]. In order to ex- tract quantitative information from such image sequences, which are helpful for understanding immune processes at cell and population levels, new software tools are needed that can accu- rately track the position of individual cells over time and in 3-dimensions for a large number of cells. Due to the complexity of this problem, existing software tools are prone to tracking errors, and often biologists must resort to manually tracking individual cells, thereby putting a practical limit on the statistics that can be obtained. Because of this, studies tend to rely only upon a small number of cell tracks and thus there is a large debate about the validity of certain biological interpretations. Therefore, a software analysis package would be an important con- tribution to the growing set of immunoinformatics tools [3]. In this paper, we describe a new set of algorithms based upon constrained particle filters as well as our software implementation for tracking a large collection of individual cells in a complex cellular environment obtained from two-photon microscopy. In a more general setting, our tool could be useful for biologists who require detailed quantitative motility analysis of many individual cells. * To whom correspondence should be addressed. Email: dnolivieri@gmail.com Copyright 2011 The Author(s). Published by Journal of Integrative Bioinformatics. This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License (http://creativecommons.org/licenses/by-nc-nd/3.0/). Journal of Integrative Bioinformatics, 8(3):180, 2011 http://journal.imbio.de doi:10.2390/biecoll-jib-2011-180 1