Abstract
Many problems in biology and medicine have a control component. Often, the goal might be to modify intracellular networks, such as gene regulatory networks or signaling networks, in order for cells to achieve a certain phenotype, what happens in cancer. If the network is represented by a mathematical model for which mathematical control approaches are available, such as systems of ordinary differential equations, then this problem might be solved systematically. Such approaches are available for some other model types, such as Boolean networks, where structure-based approaches have been developed, as well as stable motif techniques. However, increasingly many published discrete models are mixed-state or multistate, that is, some or all variables have more than two states, and thus the development of control strategies for multistate networks is needed. This paper presents a control approach broadly applicable to general multistate models based on encoding them as polynomial dynamical systems over a finite algebraic state set, and using computational algebra for finding appropriate intervention strategies. To demonstrate the feasibility and applicability of this method, we apply it to a recently developed multistate intracellular model of E2F-mediated bladder cancerous growth and to a model linking intracellular iron metabolism and oncogenic pathways. The control strategies identified for these published models are novel in some cases and represent new hypotheses, or are supported by the literature in others as potential drug targets. Our Macaulay2 scripts to find control strategies are publicly available through GitHub at https://github.com/luissv7/multistatepdscontrol.
Similar content being viewed by others
References
Albert R, Thakar J (2014) Boolean modeling: a logic-based dynamic approach for understanding signaling and regulatory networks and for making useful predictions. Wiley Interdiscip Rev Syst Biol Med 6(5):353–369
Chaouiya C, Remy E, Thieffry D (2006) Qualitative petri net modelling of genetic networks. Springer Berlin Heidelberg, Berlin, pp 95–112
Chifman J, Arat S, Deng Z, Lemler E, Pino JC, Harris LA, Kochen MA, Lopez CF, Akman SA, Torti FM et al (2017) Activated oncogenic pathway modifies iron network in breast epithelial cells: a dynamic modeling perspective. PLoS Comput Biol 13(2):e1005352
Choi M, Shi J, Jung SH, Chen X, Cho KH (2012) Attractor landscape analysis reveals feedback loops in the p53 network that control the cellular response to dna damage. Sci Signal 5(251):ra83
Correia RB, Gates AJ, Wang X, Rocha LM (2018) Cana: a python package for quantifying control and canalization in Boolean networks. Front Physiol 9
Cox D, Little J, O’shea D (1998) Using algebraic geometry, volume 185 of graduate texts in mathematics
Cox D, Little J, OShea D (2013) Ideals, varieties, and algorithms: an introduction to computational algebraic geometry and commutative algebra. Springer, Berlin
Dang CV, Reddy EP, Shokat KM, Soucek L (2017) Drugging the’undruggable’cancer targets. Nat Rev Cancer 17(8):502
Deng Z, Manz DH, Torti SV, Torti FM (2017) Effects of ferroportin-mediated iron depletion in cells representative of different histological subtypes of prostate cancer. Antioxid Redox Signal 30(8):1043–1061
Didier G, Remy E, Chaouiya C (2011) Mapping multivalued onto Boolean dynamics. J Theor Biol 270(1):177–184. https://doi.org/10.1016/j.jtbi.2010.09.017
Espinosa-Soto C, Padilla-Longoria P, Alvarez-Buylla ER (2004) A gene regulatory network model for cell-fate determination during arabidopsis thaliana flower development that is robust and recovers experimental gene expression profiles. Plant Cell 16(11):2923–2939. https://doi.org/10.1105/tpc.104.021725
Fiedler B, Mochizuki A, Kurosawa G, Saito D (2013) Dynamics and control at feedback vertex sets. i: Informative and determining nodes in regulatory networks. J Dyn Differ Equ 25(3):563–604
Gan X, Albert R (2018) General method to find the attractors of discrete dynamic models of biological systems. Phys Rev E 97(4):042308
Gates AJ, Rocha LM (2016) Control of complex networks requires both structure and dynamics. Sci Rep 6:24456
Grayson DR, Stillman ME Macaulay2, a software system for research in algebraic geometry. http://www.math.uiuc.edu/Macaulay2/
Hanahan D, Weinberg RA (2000) The hallmarks of cancer. Cell 100(1):57–70
Hanahan D, Weinberg RA (2011) Hallmarks of cancer: the next generation. Cell 144(5):646–674
Hinkelmann F, Brandon M, Guang B, McNeill R, Blekherman G, Veliz-Cuba A, Laubenbacher R (2011) Adam: analysis of discrete models of biological systems using computer algebra. BMC Bioinform 12:295. https://doi.org/10.1186/1471-2105-12-295
Hinkelmann F, Murrugarra D, Jarrah AS, Laubenbacher R (2011) A mathematical framework for agent based models of complex biological networks. Bull Math Biol 73(7):1583–1602. https://doi.org/10.1007/s11538-010-9582-8
Huang S (1999) Gene expression profiling, genetic networks, and cellular states: an integrating concept for tumorigenesis and drug discovery. J Mol Med (Berl) 77(6):469–80
Huang S, Ernberg I, Kauffman S (2009) Cancer attractors: a systems view of tumors from a gene network dynamics and developmental perspective. Semin Cell Dev Biol 20(7):869–876. https://doi.org/10.1016/j.semcdb.2009.07.003
Ireland K, Rosen M (2013) A classical introduction to modern number theory, vol 84. Springer, Berlin
Kauffman S (1969a) Homeostasis and differentiation in random genetic control networks. Nature 224(5215):177–178
Kauffman SA (1969b) Metabolic stability and epigenesis in randomly constructed genetic nets. J Theor Biol 22(3):437–67
Klamt S, Saez-Rodriguez J, Lindquist JA, Simeoni L, Gilles ED (2006) A methodology for the structural and functional analysis of signaling and regulatory networks. BMC Bioinform 7:56. https://doi.org/10.1186/1471-2105-7-56
Li R, Yang M, Chu T (2015) Controllability and observability of boolean networks arising from biology. Chaos 25(2):023104. https://doi.org/10.1063/1.4907708
Lidl R, Niederreiter H (1994) Introduction to finite fields and their applications. Cambridge University Press, Cambridge
Liu JC, Granieri L, Shrestha M, Wang DY, Vorobieva I, Rubie EA, Jones R, Ju Y, Pellecchia G, Jiang Z et al (2018) Identification of cdc25 as a common therapeutic target for triple-negative breast cancer. Cell Rep 23(1):112–126
Manz DH, Blanchette NL, Paul BT, Torti FM, Torti SV (2016) Iron and cancer: recent insights. Ann N Y Acad Sci 1368(1):149–161
Mochizuki A, Fiedler B, Kurosawa G, Saito D (2013) Dynamics and control at feedback vertex sets. ii: A faithful monitor to determine the diversity of molecular activities in regulatory networks. J Theor Biol 335:130–146
Murrugarra D, Dimitrova ES (2015) Molecular network control through boolean canalization. EURASIP J Bioinform Syst Biol 2015(1):9. https://doi.org/10.1186/s13637-015-0029-2
Murrugarra D, Veliz-Cuba A, Aguilar B, Laubenbacher R (2016) Identification of control targets in boolean molecular network models via computational algebra. BMC Syst Biol 10(1):94. https://doi.org/10.1186/s12918-016-0332-x
Naldi A, Berenguier D, Fauré A, Lopez F, Thieffry D, Chaouiya C (2009) Logical modelling of regulatory networks with ginsim 2.3. Biosystems 97(2):134–139. https://doi.org/10.1016/j.biosystems.2009.04.008
Pinnix ZK, Miller LD, Wang W, D’Agostino R, Kute T, Willingham MC, Hatcher H, Tesfay L, Sui G, Di X et al (2010) Ferroportin and iron regulation in breast cancer progression and prognosis. Sci Transl Med 2(43):43ra56–43ra56
Poret A, Boissel JP (2014) An in silico target identification using boolean network attractors: avoiding pathological phenotypes. CR Biol 337(12):661–678. https://doi.org/10.1016/j.crvi.2014.10.002
Qiu Y, Tamura T, Ching WK, Akutsu T (2014) On control of singleton attractors in multiple boolean networks: integer programming-based method. BMC Syst Biol 8 Suppl 1:S7. https://doi.org/10.1186/1752-0509-8-S1-S7
Remy E, Rebouissou S, Chaouiya C, Zinovyev A, Radvanyi F, Calzone L (2015) A modelling approach to explain mutually exclusive and co-occurring genetic alterations in bladder tumorigenesis. Cancer Res 0602
Richelle A, Lewis NE (2017) Improvements in protein production in mammalian cells from targeted metabolic engineering. Curr Opin Syst Biol 6:1–6
Saadatpour A, Wang RS, Liao A, Liu X, Loughran TP, Albert I, Albert R (2011) Dynamical and structural analysis of a t cell survival network identifies novel candidate therapeutic targets for large granular lymphocyte leukemia. PLoS Comput Biol 7(11):e1002267. https://doi.org/10.1371/journal.pcbi.1002267
Schaaf G, Hamdi M, Zwijnenburg D, Lakeman A, Geerts D, Versteeg R, Kool M (2010) Silencing of spry1 triggers complete regression of rhabdomyosarcoma tumors carrying a mutated ras gene. Cancer Res 70(2):762–771
Shmulevich I, Dougherty ER (2010) Probabilistic Boolean networks—the modeling and control of gene regulatory networks. SIAM. http://www.ec-securehost.com/SIAM/OT118.html
Slack JMW (2002) Conrad hal waddington: the last renaissance biologist? Nat Rev Genet 3(11):889–895. https://doi.org/10.1038/nrg933
Spencer-Smith R, O’Bryan JP (2017) Direct inhibition of ras: Quest for the holy grail? In: Seminars in cancer biology. Elsevier
Tan Z, Zhang S (2016) Past, present, and future of targeting ras for cancer therapies. Mini Rev Med Chem 16(5):345–357
Thieffry D, Thomas R (1995) Dynamical behaviour of biological regulatory networks-ii. Immunity control in bacteriophage lambda. Bull Math Biol 57(2):277–297. https://doi.org/10.1016/0092-8240(94)00037-D
Torti SV, Torti FM (2013) Iron and cancer: more ore to be mined. Nat Rev Cancer 13(5):342
Veliz-Cuba A, Jarrah AS, Laubenbacher R (2010) Polynomial algebra of discrete models in systems biology. Bioinformatics 26(13):1637–1643
Veliz-Cuba A, Aguilar B, Hinkelmann F, Laubenbacher R (2014) Steady state analysis of boolean molecular network models via model reduction and computational algebra. BMC Bioinform 15:221. https://doi.org/10.1186/1471-2105-15-221
Vera-Licona P, Bonnet E, Barillot E, Zinovyev A (2013) Ocsana: optimal combinations of interventions from network analysis. Bioinformatics 29(12):1571–1573. https://doi.org/10.1093/bioinformatics/btt195
Waddington CH (1957) The strategy of the genes: a discussion of some aspects of theoretical biology. Allen & Unwin, London
Yadav V, Zhang X, Liu J, Estrem S, Li S, Gong XQ, Buchanan S, Henry JR, Starling JJ, Peng SB (2012) Reactivation of mitogen-activated protein kinase (mapk) pathway by fgf receptor 3 (fgfr3)/ras mediates resistance to vemurafenib in human b-raf v600e mutant melanoma. J Biol Chem 287(33):28087–28098
Yang G, Gómez Tejeda Zañudo J, Albert R (2018) Target control in logical models using the domain of influence of nodes. Front Physiol 9:454
Yu Y, Kovacevic Z, Richardson DR (2007) Tuning cell cycle regulation with an iron key. Cell Cycle 6(16):1982–1994
Zañudo JG, Albert R (2013) An effective network reduction approach to find the dynamical repertoire of discrete dynamic networks. Chaos 23(2):025111
Zañudo JGT, Albert R (2015) Cell fate reprogramming by control of intracellular network dynamics. PLoS Comput Biol 11(4):e1004193. https://doi.org/10.1371/journal.pcbi.1004193
Zañudo JGT, Scaltriti M, Albert R (2017) A network modeling approach to elucidate drug resistance mechanisms and predict combinatorial drug treatments in breast cancer. Cancer Converg 1(1):5
Zañudo JGT, Yang G, Albert R (2017) Structure-based control of complex networks with nonlinear dynamics. Proc Natl Acad Sci USA 114(28):7234–7239. https://doi.org/10.1073/pnas.1617387114
Zañudo JGT, Yang G, Albert R (2017) Structure-based control of complex networks with nonlinear dynamics. Proc Nat Acad Sci 114(28):7234–7239
Zhan T, Rindtorff N, Boutros M (2017) Wnt signaling in cancer. Oncogene 36(11):1461
Zhang R, Shah MV, Yang J, Nyland SB, Liu X, Yun JK, Albert R, Loughran TP Jr (2008) Network model of survival signaling in large granular lymphocyte leukemia. Proc Natl Acad Sci USA 105(42):16308–16313. https://doi.org/10.1073/pnas.0806447105
Acknowledgements
Sordo Vieira is partially supported by The National Institutes of Health Grant No. 1R01AI135128-01. Laubenbacher is partially supported by The National Institutes of Health Grants No. 1R01AI135128-01, 1U01EB024501-01, and 1R01GM127909-01.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix
Appendix
In this Appendix, we denote finite fields with either \({\mathbb {F}}_q\) or \({\mathbb {F}}_p\), where p is assumed to be a prime number while q is assumed to be a power of a prime number.
1.1 Converting Mixed-State Models into Polynomial Dynamical Systems
Let q be the smallest number which is a power of a prime number such that \(q\ge |X_i|\) for all i. Consider the finite field \({\mathbb {F}}={\mathbb {F}}_q.\)
We can identify \(X_i\hookrightarrow {\mathbb {F}}\) by an injective map \(\iota _i\) for i from 1 to n. Let \(\iota =(\iota _1,\ldots , \iota _n)\). We can now consider the dynamical system \({\mathbf {F}}\) as a subsystem of a dynamical system \(\hat{{\mathbf {F}}}:{\mathbb {F}}^n\rightarrow {\mathbb {F}}^n\) as follows.
Define the map \(\alpha _i:{\mathbb {F}}\rightarrow X_i\) as \(\alpha _i(x)=\iota _i^{-1}(x)\) if \(\iota _i^{-1}(x)\ne \emptyset \) and \(\alpha _i(x)=c_i\) where \(c_i\in X_i\) otherwise. Let \(\alpha =(\alpha _1,\ldots , \alpha _n)\). Now, consider the map \(\hat{{\mathbf {F}}}=\iota \circ {\mathbf {F}} \circ \alpha \).
Notice that \(\alpha _i\) essentially “crushes” the points in \({\mathbb {F}}-\iota _i(X_i)\) into a constant in \(\iota _i(X_i)\).
Example 1
We use a slight abuse of notation for convenience: When we write an integer m here, we mean the representative in the particular finite field. Consider the sets \(X_1=\{0,1,2,\ldots ,5\},X_2=\{0,\ldots ,4\}.\)
We embed \(X_1\overset{\iota _1}{\hookrightarrow }{\mathbb {F}}_7,X_2\overset{\iota _2}{\hookrightarrow }{\mathbb {F}}_7\) by inclusion. Define \(\alpha _1, \alpha _2\) as follows.
Here \(\iota =(\iota _1, \iota _2)\), \(\alpha =(\alpha _1, \alpha _2)\), and \({\mathbb {F}}_7^2\overset{\alpha }{\rightarrow }X\overset{{\mathbf {F}}}{\rightarrow }X \overset{\iota }{\rightarrow }{\mathbb {F}}_7^2.\)
Veliz-Cuba et al. (2010) previously used a similar transformation for a finite field of prime order, \({\mathbb {F}}_p\), where the elements outside of \({\mathbb {F}}_p-\iota (X)\) were sent into the “largest” element \((p-1)\). However, in a general finite field \({\mathbb {F}}_q\), there is no adequate concept of the “largest element”. Notice that \(\hat{{\mathbf {F}}}(x_1,\ldots ,x_n)=(x_1,\ldots ,x_n)\) if and only if \((x_1,\ldots ,x_n)\) is in the image of \(\iota \) and \((\iota _1^{-1}(x_1),\ldots ,\iota _n^{-1}(x_n))\) is a fixed point of \({\mathbf {F}}.\) In particular, we can now “extend” the discrete dynamical system \({\mathbf {F}}\) to a discrete dynamical system \(\hat{{\mathbf {F}}}:{\mathbb {F}}^n \rightarrow {\mathbb {F}}^n\) without changing the dynamics of the original system.
1.2 An Approach for Deriving a Polynomial Dynamical System from a Mixed-State Dynamical System
A common approach to representing mixed-state dynamical systems is to give Boolean expressions for when a certain node will attain a given value based on the state of the other nodes (Zañudo et al. 2017; Remy et al. 2015). For example, in the signaling network model presented in Remy et al. (2015), the rule for representing how E2F3 attains values 1 or 2 are shown in Table 4.
In the case that some of the variables are Boolean (can only take one of two values), and the other variables are in a set of the same prime cardinality q, we can convert to a polynomial dynamical system over \({\mathbb {F}}_q\). If a variable \(x_i\) was Boolean to start with, we replace \(x_i\) with \(x_i^{q-1}\). For a variable, \(x_i\) that was not Boolean, we can write the polynomial representation by taking advantage of indicators functions \(q_j(x)=(\Pi _{i\in {\mathbb {F}}_q, i\ne j} (x-i))^{q-1}\) for \(j\in {\mathbb {F}}_q.\) For example, if a variable appears in a Boolean expression as \(x_i=j\), then we substitute that variable with \((\Pi _{i\in {\mathbb {F}}_q, i\ne j} (x-i))^{q-1}.\) Recall that the operator AND is equivalent to the product over \({\mathbb {F}}_2,\) the operator OR is equivalent to the operator \((x,y)\rightarrow x+y-(x+y)\) and NOT is equivalent to \(x\rightarrow 1+x\). Over \({\mathbb {F}}_q\), we define x AND y to be \((x,y)\rightarrow (x\cdot y)^{q-1}\), NOT x to be \(x\rightarrow 1-x^{q-1}\) and x OR y to be \((x,y)\rightarrow -(x\cdot y)^{q-1}+x^{q-1}+y^{q-1}.\)
Example 2
Consider the update rule for the transcription factor E2F3 from Remy et al. (2015), which takes values in the set \({\mathbb {F}}_3,\) and whose value depends on the nodes RB1, \(\text {CHECK1\_2}\), and RAS (Table 4). Here, the variables RB1 and RAS were Boolean variables, so we first substitute them with RB1\(^2\) and RAS\(^2.\) We then apply indicator functions for variables that were not Boolean. For example, \(\text {CHECK1\_2}\)=2 now becomes \(q_2(\text {CHECK1\_2})\) where \(q_2(x)=x+2\cdot x^2.\)
The final polynomial equation can now be formed by adding the individual functions together, times their respective value (Table 5).
1.3 Continuity Condition and Steady States
The continuity condition is a restriction that the state of each variable does not change by more than one unit at each time step (see, e.g., Chifman et al. 2017 for details). Intuitively, the continuity condition represents that a biological quantity cannot suddenly go from high to low (or low to high) without reaching an intermediate step. Here we show that the continuity condition on polynomial dynamical systems used in Chifman et al. (2017) does not change steady states.
Fix a prime p and consider the finite field \(k={\mathbb {F}}_p\). Fix the notation
and let \({\mathbf {F}}_i:=f_i({\mathbf {x}})\). We will always assume that the representative for \(x_i\) is in the set \(\{0,1,\ldots ,p-1\}\).
We will say that \(f:k[x_1,\ldots ,x_n]\rightarrow k^n\) is continuous if \(|x_i-f_i({\mathbf {x}})|_{{\mathbb {R}}}\in \{0,1\} \text { for } 0\le x_i\le p-1, 1\le i \le n\). Let
Any PDS \({\mathbf {F}}:k^n\rightarrow k^n\) can be made continuous by considering \(\hat{{\mathbf {F}}}:k^n\rightarrow k^n\) where \(\hat{{\mathbf {F}}}_i=h\circ ({\mathbf {F}}_i\times \pi _i)\) where \(\pi _i\) is the projection onto the ith coordinate.
Theorem 1
Let \({\mathbf {F}}:k^n\rightarrow k^n\) be a polynomial dynamical system over a finite field k and let \(\hat{{\mathbf {F}}}:k^n\rightarrow k^n\) be the polynomial dynamical system where the continuity condition has been applied to \({\mathbf {F}}\). Then the set of fixed points of \({\mathbf {F}}\), FIX(\({\mathbf {F}})\) is equal to FIX(\(\hat{{\mathbf {F}}})\)
Proof
Let \(x \in \)FIX(\({\mathbf {F}}\)), \(\pi _i:k^n\rightarrow k\) be the projection onto the ith coordinate.
Notice \(\hat{{\mathbf {F}}}_i=h\circ ({\mathbf {F}}_i\times \pi _i)\). Then \(\hat{{\mathbf {F}}}_i(x)=h\circ ({\mathbf {F}}_i\times \pi _i)(x)=h({\mathbf {F}}_i(x),x_i)=h(x_i,x_i)=x_i\).
Now, if \(x\in \text {FIX}(\hat{{\mathbf {F}}})\), we have \(h({\mathbf {F}}_i(x),x_i)=x_i\) for all i. This can only happen if \(x_i={\mathbf {F}}_i(x)\) for all i.
As a result, we have FIX\(({\mathbf {F}})=\,\)FIX(\(\hat{{\mathbf {F}}})\).
\(\square \)
Rights and permissions
About this article
Cite this article
Sordo Vieira, L., Laubenbacher, R.C. & Murrugarra, D. Control of Intracellular Molecular Networks Using Algebraic Methods. Bull Math Biol 82, 2 (2020). https://doi.org/10.1007/s11538-019-00679-w
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s11538-019-00679-w