BEGIN:VCALENDAR
VERSION:2.0
PRODID:researchseminars.org
CALSCALE:GREGORIAN
X-WR-CALNAME:researchseminars.org
BEGIN:VEVENT
SUMMARY:David F. Anderson (University of Wisconsin\, Madison (USA))
DTSTART;VALUE=DATE-TIME:20201112T160000Z
DTEND;VALUE=DATE-TIME:20201112T163000Z
DTSTAMP;VALUE=DATE-TIME:20210124T161309Z
UID:MoRN/1
DESCRIPTION:Title: Reaction network implementations of neural networks\nby
David F. Anderson (University of Wisconsin\, Madison (USA)) as part of Se
minar on the Mathematics of Reaction Networks\n\n\nAbstract\nI will give a
n overview of my recent paper with Badal Joshi and Abhishek Deshpande\, wh
ich is entitled "On reaction network implementations of neural networks."
In particular\, I will show how reaction networks can be constructed that
"implement" a given neural network. I will also detail our theoretical r
esults\, which prove that the ODEs associated with certain reaction networ
k implementations of neural networks have desirable properties including (
i) existence of unique positive fixed points that are smooth in the parame
ters of the model (necessary for gradient descent)\, and (ii) fast converg
ence to the fixed point regardless of initial condition (necessary for eff
icient implementation). I'll start the talk with a brief primer on neural
networks\, but will assume familiarity with reaction networks.\n
END:VEVENT
BEGIN:VEVENT
SUMMARY:Beatriz Pascual Escudero (Universidad Carlos III (Spain))
DTSTART;VALUE=DATE-TIME:20201203T160000Z
DTEND;VALUE=DATE-TIME:20201203T163000Z
DTSTAMP;VALUE=DATE-TIME:20210124T161309Z
UID:MoRN/2
DESCRIPTION:Title: Necessary conditions for ACR in Reaction Networks\nby B
eatriz Pascual Escudero (Universidad Carlos III (Spain)) as part of Semina
r on the Mathematics of Reaction Networks\n\n\nAbstract\nA biological syst
em has absolute concentration robustness (ACR) for some molecular species
if the concentration of this species does not vary among the different ste
ady states that the network admits. In particular\, this concentration is
independent of the initial conditions. This interesting feature confers th
e system a highly desirable property in order to adapt to environmental co
nditions\, which makes it useful\, for instance\, in synthetic biology. Wh
ile some classes of networks with ACR have been described (Shinar and Fein
berg 2010\; Karp et al. 2012)\, as well as some techniques to check a netw
ork for ACR (Pérez Millán 2011\; Kuwahara et al. 2017)\, finding network
s with this property is a difficult task in general.\n\nMotivated by this
problem\, we studied local and global notions of robustness on the set of
(real positive) solutions of a system of polynomial equations\, and in par
ticular on the set of steady states of a reaction network. Algebraic geome
try allowed us to provide a practical test on necessary conditions for ACR
. Properties of real and complex algebraic varieties are necessary for the
results\, while the test ends up being a linear algebra computation.\n
END:VEVENT
BEGIN:VEVENT
SUMMARY:Casian Pantea (West Virginia University (USA))
DTSTART;VALUE=DATE-TIME:20201203T163000Z
DTEND;VALUE=DATE-TIME:20201203T170000Z
DTSTAMP;VALUE=DATE-TIME:20210124T161309Z
UID:MoRN/3
DESCRIPTION:Title: Inheritance of Hopf bifurcations in reaction networks\n
by Casian Pantea (West Virginia University (USA)) as part of Seminar on th
e Mathematics of Reaction Networks\n\n\nAbstract\nInspired by recent work
on multistationarity\, we consider the question: "when can we conclude tha
t a network admits Hopf bifurcations if one of its subnetworks has them?
” In particular\, we analyze a number of operations on reaction networks
(like adding certain reactions\, or adding inflows/outflows) that may pr
eserve Hopf bifurcations as we build up the network . This is joint work w
ith C.Conradi\, A. Dickenstein\, and M. Mincheva.\n
END:VEVENT
BEGIN:VEVENT
SUMMARY:Lea Popovic (Concordia University)
DTSTART;VALUE=DATE-TIME:20201112T163000Z
DTEND;VALUE=DATE-TIME:20201112T170000Z
DTSTAMP;VALUE=DATE-TIME:20210124T161309Z
UID:MoRN/4
DESCRIPTION:Title: A spatially heterogeneous stochastic model for chemical
reaction networks\nby Lea Popovic (Concordia University) as part of Semin
ar on the Mathematics of Reaction Networks\n\n\nAbstract\nI will present a
measure-valued framework for stochastic modelling of chemical reaction ne
tworks with spatial heterogeneity. Reactions rates at a spatial location a
re proportional to the mass of different species present locally\, and to
a location specific chemical rate that is allowed to be a function of the
local or global mass of different species. The benefit of the framework is
in rigorous approximation limits that exploit multi-scale aspects of the
system. When the mass of all species scales the same way\, we get classica
l deterministic limit described by PDEs. When the mass of some species in
the scaling limit is discrete while the mass of the others is continuous\,
we obtain a new type of spatial random evolution process in which discret
e mass evolves stochastically and the continuous mass evolves according to
PDEs between consecutive jump times of the discrete part. Some useful pro
perties of the limiting process are inherited from the pre-limiting sequen
ce\, and could be used in devising simulation algorithms.\n\nThis is joint
work with Amandine Veber (Paris V\, Polytechnique-Saclay)\n
END:VEVENT
BEGIN:VEVENT
SUMMARY:Nida Obatake (Texas A&M (USA))
DTSTART;VALUE=DATE-TIME:20201210T160000Z
DTEND;VALUE=DATE-TIME:20201210T163000Z
DTSTAMP;VALUE=DATE-TIME:20210124T161309Z
UID:MoRN/5
DESCRIPTION:Title: Mixed volume of reaction networks\nby Nida Obatake (Tex
as A&M (USA)) as part of Seminar on the Mathematics of Reaction Networks\n
\n\nAbstract\nAn important invariant of a chemical reaction network is its
maximum number of positive steady states. This number\, however\, is in g
eneral difficult to compute. We introduce an upper bound on this number—
namely\, a network’s mixed volume — that is easy to compute. We show
that\, for certain biological signaling networks\, the mixed volume does n
ot greatly exceed the maximum number of positive steady states. We investi
gate this overcount and also compute the mixed volumes of small networks (
those with only a few species or reactions). Joint work with Anne Shiu\, D
ilruba Sofia\, Angelica Torres\, and Xiaoxian Tang.\n
END:VEVENT
BEGIN:VEVENT
SUMMARY:Ankit Gupta (ETHZ (Switzerland))
DTSTART;VALUE=DATE-TIME:20201210T163000Z
DTEND;VALUE=DATE-TIME:20201210T170000Z
DTSTAMP;VALUE=DATE-TIME:20210124T161309Z
UID:MoRN/6
DESCRIPTION:Title: Frequency Spectra and the Color of Cellular Noise\nby A
nkit Gupta (ETHZ (Switzerland)) as part of Seminar on the Mathematics of R
eaction Networks\n\nAbstract: TBA\n
END:VEVENT
BEGIN:VEVENT
SUMMARY:Polly Yu (University of Wisconsin\, Madison)
DTSTART;VALUE=DATE-TIME:20210114T160000Z
DTEND;VALUE=DATE-TIME:20210114T163000Z
DTSTAMP;VALUE=DATE-TIME:20210124T161309Z
UID:MoRN/7
DESCRIPTION:Title: Dynamically Equivalent Mass-Action Systems: A Survey of
Recent Results\nby Polly Yu (University of Wisconsin\, Madison) as part o
f Seminar on the Mathematics of Reaction Networks\n\n\nAbstract\nUnder mas
s-action kinetics\, each reaction network uniquely gives rise to a system
of ODEs. However\, the converse is not true\; for a given system of ODEs k
nown to come from a mass-action systems\, there are many reaction networks
that serve as a candidate. In this talk\, I will introduce the notion of
dynamical equivalence\, emphasize a convenient way of thinking about it\,
and survey some recent results on dynamical equivalence to complex-balance
d or detailed-balanced systems.\n
END:VEVENT
BEGIN:VEVENT
SUMMARY:Carsten Wiuf (University of Copenhagen)
DTSTART;VALUE=DATE-TIME:20210225T160000Z
DTEND;VALUE=DATE-TIME:20210225T163000Z
DTSTAMP;VALUE=DATE-TIME:20210124T161309Z
UID:MoRN/8
DESCRIPTION:by Carsten Wiuf (University of Copenhagen) as part of Seminar
on the Mathematics of Reaction Networks\n\nAbstract: TBA\n
END:VEVENT
BEGIN:VEVENT
SUMMARY:Jinsu Kim (UC Irvine)
DTSTART;VALUE=DATE-TIME:20210128T160000Z
DTEND;VALUE=DATE-TIME:20210128T163000Z
DTSTAMP;VALUE=DATE-TIME:20210124T161309Z
UID:MoRN/9
DESCRIPTION:Title: Identifiability of Stochastically Modelled Reaction Net
works\nby Jinsu Kim (UC Irvine) as part of Seminar on the Mathematics of R
eaction Networks\n\n\nAbstract\nWhen an underlying reaction network is giv
en for a biochemical system\, the system dynamics can be modeled with vari
ous mathematical frameworks such as continuous-time Markov processes. In t
his manuscript\, the identifiability of the underlying network structure w
ith a given stochastic system dynamics is studied. It is shown that some d
ata types related to the associated stochastic dynamics can uniquely ident
ify the underlying network structure as well as the system parameters. The
accuracy of the presented network inference is investigated when given dy
namical data is obtained via stochastic simulations.\n
END:VEVENT
BEGIN:VEVENT
SUMMARY:Peter Szmolyan (TU Wien)
DTSTART;VALUE=DATE-TIME:20210325T163000Z
DTEND;VALUE=DATE-TIME:20210325T170000Z
DTSTAMP;VALUE=DATE-TIME:20210124T161309Z
UID:MoRN/10
DESCRIPTION:Title: TBA (Dimension reduction in chemical systems with sever
al slow manifolds)\nby Peter Szmolyan (TU Wien) as part of Seminar on the
Mathematics of Reaction Networks\n\nAbstract: TBA\n
END:VEVENT
BEGIN:VEVENT
SUMMARY:Elisa Tonello (Freie Universität\, Berlin)
DTSTART;VALUE=DATE-TIME:20210211T160000Z
DTEND;VALUE=DATE-TIME:20210211T163000Z
DTSTAMP;VALUE=DATE-TIME:20210124T161309Z
UID:MoRN/11
DESCRIPTION:by Elisa Tonello (Freie Universität\, Berlin) as part of Semi
nar on the Mathematics of Reaction Networks\n\nAbstract: TBA\n
END:VEVENT
BEGIN:VEVENT
SUMMARY:Noel Fortun (De La Salle University Manila)
DTSTART;VALUE=DATE-TIME:20210211T163000Z
DTEND;VALUE=DATE-TIME:20210211T170000Z
DTSTAMP;VALUE=DATE-TIME:20210124T161309Z
UID:MoRN/12
DESCRIPTION:Title: TBA (Concentration robustness in PLK and PYK)\nby Noel
Fortun (De La Salle University Manila) as part of Seminar on the Mathemati
cs of Reaction Networks\n\nAbstract: TBA\n
END:VEVENT
BEGIN:VEVENT
SUMMARY:Justin Eilertsen (University of Michigan)
DTSTART;VALUE=DATE-TIME:20210114T163000Z
DTEND;VALUE=DATE-TIME:20210114T170000Z
DTSTAMP;VALUE=DATE-TIME:20210124T161309Z
UID:MoRN/13
DESCRIPTION:Title: The current state of quasi-steady-state approximations:
manifolds\, time scales\, singularities\, and stochastic fluctuations\nby
Justin Eilertsen (University of Michigan) as part of Seminar on the Mathe
matics of Reaction Networks\n\n\nAbstract\nOver the past decade\, mathemat
icians have made considerable progress concerning the theory and\napplicab
ility of quasi-steady-state (QSS) approximations in chemical kinetics. The
application of Fenichel theory has revealed that QSS reduction in chemica
l kinetics is far richer than previously thought\, even in low-dimensional
systems that do not exhibit oscillatory behavior. In this talk\, I will d
iscuss recent discoveries that have emerged in the \nfield of mathematical
enzyme kinetics\, including methodologies for obtaining perturbation para
meters\, singular points\, dynamic bifurcations and scaling laws. If time
permits\, I will also discuss the applicability of QSS reductions in stoch
astic environments\, and comment on some open problems in both determinist
ic and stochastic enzyme kinetics.\n
END:VEVENT
BEGIN:VEVENT
SUMMARY:Mark Curiel (University of Hawaii at Manoa)
DTSTART;VALUE=DATE-TIME:20210128T163000Z
DTEND;VALUE=DATE-TIME:20210128T170000Z
DTSTAMP;VALUE=DATE-TIME:20210124T161309Z
UID:MoRN/14
DESCRIPTION:Title: When do two networks have the same steady-state ideal?\
nby Mark Curiel (University of Hawaii at Manoa) as part of Seminar on the
Mathematics of Reaction Networks\n\n\nAbstract\nUnder the assumption of ma
ss action kinetics\, the associated dynamical system of a reaction network
is polynomial. We consider the ideals generated by these polynomials\, wh
ich are called steady-state ideals. Steady-state ideals appear in multiple
contexts within the chemical reaction network literature\, however they h
ave yet to be systematically studied. To begin such a study\, we ask and p
artially answer the following question: when do two reaction networks give
rise to the same steady-state ideal? In particular\, our main results des
cribe three operations on the reaction graph that preserve the steady-stat
e ideal. Furthermore\, since the motivation for this work is the classific
ation of steady-state ideals\, monomials play a primary role. To this end
\, combinatorial conditions are given to identify monomials in a steady-st
ate ideal\, and we give a sufficient condition for a steady-state ideal to
be monomial.\n
END:VEVENT
END:VCALENDAR