gallery/exp_acoustics
gallery/sim_acoustics
gallery/untitled
gallery/untitled copy

Simulation vs experiments of vocalization in a long tube

Immune-inspired ultra-dense HetNets

Bio-inspired algorithms

 

Design of efficient decentralized decision-making algorithms for the control of distributed multi-agent systems is an open engineering problem. Currently, there is no state-of-the-art methodology that allows for the design of such agent decision-making algorithms that ensure the system’s robust self-organized behavior. Solution ideas are expected from biological multicellular systems where cells decide about their phenotype based on local interactions.

 

Bio-inspired self-organized networks

The project aims at the design of bio-inspired algorithms for the regulation of self-organized heterogeneous wireless networks (HetNets) that effectively satisfy the demands of mobile phone users. The self-regulation of base station on/off switching in ultra-dense small cell networks (type of HetNets), based on spatial fluctuations of users’ traffic demand, is still a major challenge. In collaboration with engineers, we develop an agent-based model that transfers ideas from the phenotypic regulation of the immune system to self-organization of HetNets. In particular, different immune cell phenotypes correspond to different types of antennas and infected cells correspond to mobile phone users. A set of analogies between mechanisms in the immune system and cellular wireless networks has been already established. Initial simulations provide a successful proof of concept [A23, E1].

 

Bioacoustics

 

Voice production is parted of the superposition two procedures: Phonation and Articulation. The first deals with the vibrations of vocal folds and the output feed the vocal tract [A2]. The different geometries of it, during speech, define the articulations. The acoustical signal is the output of the source signal filtered by the vocal tract. The interactions of source-filter may lead to every possible dynamical phenomenon, like steady states, period doublings, quasi-periodic oscillations and chaotic vibrations. Moreover various illnesses of the vocal system can also let the creation of such phenomena. A classification and a clarification of the underlying mechanisms can create a mapping between phenomena-mechanism, leading to the source of the malfunction. In addition, the opera improvisers are able to perform amazing voice acrobatics-registers.

Last but not least, the above phenomena are appearing often in animal vocalization. The extreme geometries and the creativity of nature in discovering mechanisms offer a lot of material to study. The amazing singings of birds, the strange modulation of the formants (resonant frequencies) of Diana monkey in leopard alarm call, the evolutionary reason of air sacks are open questions.  Of particular interest are the implications of vocalization mechanics on the evolutionary dynamics of from animal vocalizations to human voice [A1A4].

B. Physiological and pathological development of multicellular systems
B1. Impact of cell phenotypic plasticity brain tumor growth and invasion

Cancer cell adaption to microenvironmental changes is a major cause of cancer therapy failure. It is unclear how phenotypic plasticity contributes to brain tumor invasion and progression. In particular I have focused on the plasticity between migratory and proliferative phenotypes, the so-called Go-or-Grow, and studied its impact on brain tumor dynamics. With the use of mathematical modeling and analysis, I studied the impact of Go-or-Grow on tumor growth and invasion. My work has shown that migration/proliferation dichotomy is a key factor for glioblastoma tumor recurrence after surgery and contributes to the emergence of brain tumor robustness against therapies [A9A12A14A15A19A21]. Currently, the focus was on the impact of tumor-induced vascularization dynamics - especially on the phenomenon of vaso-occlusion – on brain tumor invasion. Moreover, with the help of A. Cavalcanti we try to understand the role of adhesion forces in epithelial-mesenchymal transition (EMT) using our M^3 approach.

 

B2. Cell fate decision-making in retinal development

Vertebrate retina constitutes part of the Central Nervous System (CNS). Its development requires the hierarchical regulation of a series of cell decision making processes. Among these is cell cycle exit, cell migration as well as the activation of multiple regulatory pathways which involve transcription factors and cell-cell signaling events. Surprisingly, the plethora of heterogeneous retina cell types develop from a common pool of progenitor cells. However, in contrast to other well studied parts of the CNS (neocortex organization), retinal cell diversification is proposed to be achieved by the sequential production of cell types based on the dynamic interplay of intrinsic and extrinsic factors [Cepko2014]. Our approach focusing on two key events in retinal development based on experimental data provided by our collaborator Prof. M. Perron (CNRS, Paris): (i) differentiation of ciliary marginal zone stem cell and (ii) proliferation of apical/subapical progenitors in central retina development. We developed the first multiscale spatiotemporal mathematical model to investigate the effects of intrinsic (e.g. symmetric/asymmetric divisions) and extrinsic (e.g. external signals) factors in retinal cell fate determination. In particular, our results suggested that the robustness of apical versus sub-apical development relies on the dynamic interplay between apical to sub-apical symmetric to asymmetric division as well as on communication with cell local microenvironment [A22]. Even though many of the components regarding differentiation programs are known, how noise may affect cell fate specification over time remains poorly understood. In this regard, we have identified that extrinsic noise in cell fate specification signaling (mediated by Notch-Delta) [Perron and Harris, 2000] may force cellular sub-populations to a metastable, transient, frustrated, mixed “state = fate” [A20]. Further design ideas for de-noising mechanisms of the proposed model have been developed [B5].

 

B3. Immune system plasticity in tumors and adipose tissue

B3.1 Tumor-immune interactions

Cancer cell adaption to microenvironmental/stromal changes in combination with immune cell decisions is crucial to tumor prognosis and therapeutic success. The complex interplay of two highly plastic systems requires the combination of biomedical experiments and mathematical modelling. Of particular interest is the impact of common determinant factors for both systems such as vascularization. Nevertheless, despite years of research, there are still many unanswered questions regarding the critical components of the tumor microenvironment that play important roles in the immuno-oncological mechanisms. We develop mathematical models to verify theoretical assumptions and suggest modifications of existing theories, as well as to perform in silico experiments to gain insights in tumor-immune interactions [A25].

 

B3.2 Macrophage plasticity in breast tumor development

Tumor associated macrophages (TAMs) are proven to be central to the prognosis of tumor fate. Macrophages exhibit a remarkable ability in changing between two extreme phenotypes M1 and M2, exhibit anti-tumoral and pro-tumoral behaviors, respectively. We use mathematical modeling to study, in combination with biomediclal data provided from Prof. F. Feuerhacke (MH Hannover), the dynamics of tumor cell-macrophages system, in the context of breast cancer. In particular, we envisage understanding how cancer hijacks macrophage functions and generates a tumor-promoting microenvironment will be translated into innovative biomarker analysis workflows with the potential to broaden applications to other cancer types.

 

B3.3 Macrophage plasticity in adipose tissue development

The dynamic interplay between metabolic disorders such as obesity and immunity has emerged as a major research focus in recent year. As a multitude of biochemical and biophysical metabolic pathways (mTOR) can interfere with immune cells (macrophages and Tregs) and the metabolic regulation of immune cell function, it remains a challenge to unravel the complexity of interacting signal and processes and turn this knowledge into more efficacious therapies. In collaboration with one of the leading medical experts in this field, Prof. T. Chavakis, we develop multiscale spatio-temporal models that can help to elucidate the complex interplay between inflammatory and metabolic pathways by providing a quantitative framework for testing the research hypotheses

 

C. Bio-inspired algorithms

Design of efficient decentralized decision-making algorithms for the control of distributed multi-agent systems is an open engineering problem. Currently, there is no state-of-the-art methodology that allows for the design of such agent decision-making algorithms that ensure the system’s robust self-organized behavior. Solution ideas are expected from biological multicellular systems where cells decide about their phenotype based on local interactions.

 

C1. Bio-inspired self-organized networks

The project aims at the design of bio-inspired algorithms for the regulation of self-organized heterogeneous wireless networks (HetNets) that effectively satisfy the demands of mobile phone users. The self-regulation of base station on/off switching in ultra-dense small cell networks (type of HetNets), based on spatial fluctuations of users’ traffic demand, is still a major challenge. In collaboration with engineers, we develop an agent-based model that transfers ideas from the phenotypic regulation of the immune system to self-organization of HetNets. In particular, different immune cell phenotypes correspond to different types of antennas and infected cells correspond to mobile phone users. A set of analogies between mechanisms in the immune system and cellular wireless networks has been already established. Initial simulations provide a successful proof of concept [A23, E1].

 

D. Bioacoustics

Voice production is parted of the superposition two procedures: Phonation and Articulation. The first deals with the vibrations of vocal folds and the output feed the vocal tract [A2]. The different geometries of it, during speech, define the articulations. The acoustical signal is the output of the source signal filtered by the vocal tract. The interactions of source-filter may lead to every possible dynamical phenomenon, like steady states, period doublings, quasi-periodic oscillations and chaotic vibrations. Moreover various illnesses of the vocal system can also let the creation of such phenomena. A classification and a clarification of the underlying mechanisms can create a mapping between phenomena-mechanism, leading to the source of the malfunction. In addition, the opera improvisers are able to perform amazing voice acrobatics-registers.

Last but not least, the above phenomena are appearing often in animal vocalization. The extreme geometries and the creativity of nature in discovering mechanisms offer a lot of material to study. The amazing singings of birds, the strange modulation of the formants (resonant frequencies) of Diana monkey in leopard alarm call, the evolutionary reason of air sacks are open questions.  Of particular interest are the implications of vocalization mechanics on the evolutionary dynamics of from animal vocalizations to human voice [A1A4].