This page contains an implementation in Maple of the Markov chain models of chromosomal instability that we introduced in the papers:
- A. Laughney, S. Elizalde, G. Genovese and S. Bakhoum, Dynamics of tumor heterogeneity derived from clonal karyotypic evolution, Cell Rep. 12 (2015), 809-820.
- S. Elizalde, A. Laughney and S. Bakhoum, A Markov chain for numerical chromosomal instability in clonally expanding populations, PLoS Comput. Biol. 14(9): e1006447.
- T. Watkins, E. Lim, M. Petkovic, S. Elizalde, N. Birkbak, G. Wilson, A. Rowan, S. Dewhurst, J. Demeulemeester, S. Dentro, S. Horswel, L. Au, K. Haase, M. Escudero, W. Gronroos, R. Rosenthal, M. Al Bakir, H. Xu, K. Litchfield, W.T. Lu, T. Mourikis, M. Dietzen, L. Spain, G. Cresswell, D. Biswas, P. Lamy, I. Nordentoft, K. Harbst, F. Castro-Giner, L. Yates, F. Caramia, F. Jaulin, C. Vicier, I. Tomlinson, R. Cho, B. Bastian, L. Dyrskjot, G. Jonsson, P. Savas, S. Loi, P. Campbell, F. Andre, N. Luscombe, Z. Szallasi, M. Jamal-Hanjani, S. Turajlic, P. Van Loo, S. Bakhoum, R. Schwarz, N. McGranahan and C. Swanton, Pervasive Chromosomal Instability and Karyotype Order During Tumour Evolution, Nature 587 (2020), 126-132.
Maple files
- model_with_scores.mw contains an implementation of the Markov chain model from the first two papers above. It includes all the necessary functions to run model with and without chromosome scores, as well as functions used to produce the plots in the paper showing distributions of karyotypes and related data, along with examples of their usage.
- model_with_scores_and_arm_level_events.mw
contains an implementation of the variation of the model used in the third paper above, which incorporates arm-level events and keeps track of copy numbers of each arm.
The first part of the file contains all the necessary functions to run the model. The second part runs the model on different inputs and different parameter values, and it outputs the model predictions in separate files. The third part imports the data from the four files below provided by Tom Watkins, containing tumor multiregion genomic sequence data and chromosome copy number information from 1421 samples from 394 tumours across 25 cancer types. The fourth part compares the model predictions to the patient data, and produces plots that measure the prediction error of different versions of the model with different parameter values, in order to compare their performance.