VDNA-Lab: A Computational Simulation Platform for DNA
Multi-strand Dynamics
Frankie Spencer
1 a
, Usman Sanwal
1 b
and Eugen Czeizler
1,2 c
1
Department of Information Technologies,
˚
Abo Akademi University, Turku, Finland
2
National Institute for Research and Development of Biological Sciences, Bucharest, Romania
Keywords:
DNA Assembly Simulator, Multi-strand Assembly, Rule-based Modeling, Graphical User Interface,
Course-grained Modeling.
Abstract:
The dynamics of nucleic-acids dynamical systems is intrinsically based on local interaction. The major acting
mechanisms are that of Watson-Crick complementarity on one-hand, generating binding events, and ther-
mal energy on the other, generating random motion and un-binding. It is thus predictable that such systems
can be successfully captured by computational modeling paradigms based on local interactions, such as the
rule-based modeling methodology. In this research we introduce the Virtual DNA Lab (VDNA-Lab) a sim-
ulation tool which provides an easy to use graphical interface for creating, running and visualizing synthetic
simulations for DNA assembly systems, such as assembly of DNA nanostructures, strand displacement cas-
cades systems, DNA-tile assembly etc. It employs a custom designed model, implemented in the BioNetGen
Language (BNGL) formalism, to capture the DNA dynamics, as well as the NFsim computational model-
ing engine to run simulations and generate outputs. These outputs can be visualized using the VDNA-Lab’s
own visualization tool, which allows also for further analysis and filtering. The software is freely available at
https://github.com/Frankie-Spencer/virtual dna lab.
1 INTRODUCTION
The fast increase of computational power experienced
in the past decade has made it possible for computer
scientists to develop various modeling and simula-
tion techniques giving them access to very high levels
of details for various self-assembly phenomena, de-
spite that achieving such levels is prohibitively expen-
sive in real experiments. Thus, the use of computer
simulations in the field of self-assembly is now well
accepted as well as its role in better understanding
of many quantitative details of such systems(Thomas
and Schwartz, 2017), (Kamerlin and Warshel, 2011),
(Glotzer et al., 2004). A major challenge for the mod-
eling of self-assembly biological phenomena lies in
the fact that we have to deal with an extremely large
space of possible assembly pathways accessible to the
intermediate species of a self-assembly reaction net-
work. Another challenge arises from the fact that
self-assembly reactions are sensitive to many physical
constrains (e.g. structural, chemical) available within
a
https://orcid.org/0000-0002-6775-1832
b
https://orcid.org/0000-0002-2178-3329
c
https://orcid.org/0000-0002-1607-1554
the original in vivo environment, conditions which
are hard to be replicated within the in silico computa-
tional models (Ariga and et al., 2008). Despite these
challenges there are a variety of modeling methods
available such as SSA/Gillespie approaches (Sweeney
and et al., 2008), (Blinov and et al., 2004) or Rule-
based approaches (Mohammed et al., 2017), (Danos
and et al., 2009), which have been proven valuable for
the study and analysis of self-assembly systems.
Nanotechnology is one of the leading worldwide
undertakings of the scientific community. Within this
field, DNA has proved itself as a very powerful and
versatile construction material(Kuzyk et al., 2009),
(Li and et al., 2011) and (Lund and et al., 2010).
Recent advances in DNA-based nanotechnology have
opened the way towards the systematic engineering
of inexpensive nanoscale devices for a multitude of
purposes (Krishnan and Simmel, 2011), (Li and et al.,
2011), (Lund and et al., 2010). Self-assembly prop-
erties of DNA, which are based on specific and pre-
dictable Watson-Crick based pairing rules, make it a
highly malleable and controllable nanomaterial (See-
man, 2003).
There are several tools available for modeling and
288
Spencer, F., Sanwal, U. and Czeizler, E.
VDNA-Lab: A Computational Simulation Platform for DNA Multi-strand Dynamics.
DOI: 10.5220/0010546302880294
In Proceedings of the 11th International Conference on Simulation and Modeling Methodologies, Technologies and Applications (SIMULTECH 2021), pages 288-294
ISBN: 978-989-758-528-9
Copyright
c
2021 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
visualization of complex DNA-based self-assembly
constructions. These tools include the Visual DSD
(DNA Strand Displacement) tool (Lakin and et al.,
2011), focused on modeling DNA strand displace-
ment reaction systems, Iowa State University Tile As-
sembly Simulator (ISU TAS) (Patitz, 2011), focused
on modeling DNA-tile assembly, or oxDNA (Popple-
ton and et al., 2020), which is a course grained mod-
eling tool for the physical atom-level simulation of
DNA assembly systems. Rule-based modeling ap-
proaches have also been used for modeling the ki-
netics of DNA-based Tile Assembly Systems (Mo-
hammed et al., 2017), as well as for DNA strand dis-
placement cascades (Gautam. et al., 2020).
In our work, we provide a framework for rule-
based modeling of generalized DNA multi-strand dy-
namics, where DNA molecules and their interac-
tions are modeled at nucleotide level. We have used
the BioNetGen Language (BNGL) formalism (Blinov
and et al., 2004) and the NFsim computational plat-
form (Sneddon and et al., 2011) as well as a com-
prehensive collection of aiding (Python) subroutines
which facilitate the user interface and make the mod-
eling experience available for the user without having
any prior knowledge of the BNGL modeling frame-
work. In order to track and report global mapping
of the components within a heterogeneous complex,
we have developed a visualization interface, details
provided in Section 3, which not only helps to vi-
sualize the results but also can be used for further
numerical analysis. The software is freely available,
see (Spencer et al., 2021).
2 MATERIAL AND METHODS
2.1 Rule-based Modeling
Modeling and simulation of self-assembly biochemi-
cal systems is a challenging computational task, as the
modeler has to deal with both a high combinatorial
number of species, namely all the generated partial-
and sub-assemblies, and also with a large numbers
of reactions between these species. The Rule-based
modeling approach, which employs the idea of sub-
dividing molecules, or generally species-entities, into
their primary components, denoting protein domains,
active sites or any other feature of the particle, can
be used to solve this problem (Faeder et al., 2009).
Within the rule-based modeling paradigm, molecules
are represented as agents with a finite number of inde-
pendent sites. The sites provide the agent-agent bind-
ing functionality and as a result molecular complexes
are generated. Rules are only defined on the basis of
local patterns which provide a compact representa-
tion about agents’ interaction. In rule-based model-
ing, models are defined in terms of a specific syntax
(rules) instead of writing the model mathematically
directly. After that a semantic is presented in order to
cover the differences between what is written and the
mathematical definition of its computation. A care-
fully described syntax increases the model’s accessi-
bility for domain experts that are not familiar with the
modeling and mathematical formalisms (Faeder et al.,
2009).
Two useful simulation tools for rule-based mod-
eling are BioNetGen (Faeder et al., 2009) and NF-
sim (Sneddon and et al., 2011). BioNetGen is a
BNGL-compatible simulation tool that implements
various indirect methods including both determin-
istic and stochastic methods. NFsim is also a
BNGL-compatible simulation tool that implements a
(stochastic) direct method which applies a specific
network-free simulation and thus avoids enumeration
of species and reactions, that may be intractable for
large models (Yang and Hlavacek, 2011). There are
six different blocks of a BNGL file, namely: param-
eters, species, observables, molecule types, reaction
rules and functions. Further detail about these blocks
and NFSim simulator can be found in (Faeder et al.,
2009) and (Sneddon and et al., 2011).
2.2 A Computational Modeling
Framework for DNA Multi-strand
Dynamics
We have created a generic computational modeling
framework for DNA multi-strand dynamics using the
BioNetGen Language (BNGL) formalism and the
NFsim computational platform. The DNA molecules
and the subsequent interactions are captured starting
from nucleotide level. Each nucleotide (as a bio-
entity) is represented as an individual instance of a
generic agent N (from nucleotide) which has 1 state-
characteristic site, b (from base), and 3 connecting
sites: 5’, 3’, and W. Relating to its biological coun-
terpart, the site b can be in exactly one of the four
possible states: A (Adenine), G (Guanine), C (Cyto-
sine), or T (Thymine); moreover, once initialized, site
b will never change its state. The remaining sites of
agent N are used for connecting the nucleotide within
the single-stranded molecule it is part of
1
, i.e., us-
1
DNA is composed primarily of nucleotides with a sugar-
phosphate backbone; the nucleotides are attached to the
backbone to form a structure which is known as a single-
stranded DNA (ssDNA) molecule. The ssDNA is repre-
sented with two different ends. By convention, one end of
the backbone is know as the 5’ end and the second one is
VDNA-Lab: A Computational Simulation Platform for DNA Multi-strand Dynamics
289
ing sites 5’ and 3’, and for connecting the nucleotide
with a complementary base-pair
2
, using site W. For
example, within our rule-based model, the basic struc-
ture for the single-stranded DNA (ssDNA) molecule
ATTGCTA is shown in Figure 1.
Figure 1: The schematic representation of the ssDNA com-
plex ATTGCTA within our BNGL model.
VDNA-Lab takes as input populations of isolated
(5’-to-3’ oriented) single-stranded DNA molecules
(ssDNAs) and simulates their binding and dissocia-
tion reactions. At the core of this model implemen-
tation lie 12 binding and un-binding local interaction
rules, each with its own kinetic rate constant, and each
implemented through one or several rule-based reac-
tions. Using these reactions/rules, we capture the dy-
namics of the DNA-dynamical system, by modeling:
the initial binding of short toeholds, i.e., short
complementary sequences of 3-to-7 nucleotides;
the ”breathing” dynamics in-between bounded ss-
DNAs, i.e. zipping and unzipping reactions;
random un-binding events (between one pair of
nucleotides);
the un-binding of loosely connected ssDNAs.
The list of all rule-types and their associated ki-
netic rate constants is presented in Table 1. By de-
fault, VDNA-Lab provides some normalized values
for these kinetic rate constants, see the 4-th column
of the table. However, any of these values can be up-
dated by the user using the Advanced option.
3 RESULTS: THE VDNA-Lab
SOFTWARE
We have developed the VDNA-Lab (Virtual DNA
Lab) software to provide an easy to use graphical in-
terface for creating, running and visualizing synthetic
simulations for DNA assembly systems, such as as-
sembly of DNA nanostructures, strand displacement
cascades systems, DNA-tile assembly etc. It uses the
NFsim computational modeling engine to run simu-
lations and generate outputs. These outputs can be
visualized using the VDNA-Lab’s visualization tool,
known as the 3’ end.
2
The base-pair nucleotides A-T and C-G are complemen-
tary which means their shape allows them to bond together
with hydrogen bonds.
which allows also some further analysis of the simu-
lation output. Here we explain the different features
and parameters of the VDNA-Lab software.
The tool consists of three main functionalities,
each reachable from a specific tab on the upper left
corner of the main window:
Create Test Tube for creating new test tube input
files; these are called also species files, as they are
stored on computer with the extension .species
Run Experiment for simulating the assembly pro-
cess within a given input test tube file, and
Visualize Test Tube for visualizing the content of
a test tube at the beginning, middle, or end of an
experiment.
The Create Test Tube feature of VDNA-Lab, see
Fig. 2 is used to create/edit new species files, to be
used as input in the Run Experiment process. ssDNA
sequences are entered by the user (only strings over
A,T,C and G are allowed to be entered as input) in
the Desired ssDNA sequence tab and the number of
copies of the generated ssDNA structure has to be in-
troduced in the Amount tab (by default 1 copy is cre-
ated). Once the user presses the Add button, these
sequences are added to the List of ssDNA to be cre-
ated text box. After the ssDNA sequences are cre-
ated, the user can Edit, Delete, or Reset all of them.
The user can also import the ssDNA sequences, or
even more complicated multi-ssDNA complexes, us-
ing the Import from Species button on the right hand
side. Such species files can be extracted for exam-
ple from the default output files generated during the
numerical simulation phase, i.e., from the ”Run ex-
periment” tab. Thus, the user can select the input
3
single complementary bases
4
The two interacting ssDNA segments can be part of the
same ssDNA molecule, i.e., a hairpin loop, or be two dis-
tinct ssDNA molecules bound within the same complex.
5
That is, both ssDNA strands continue with pairwise non-
bounded sequences to one direction -at least one strand
should not be bound to anything else- and bound sequences
of length at least 1 to the other direction
6
There are several cases within the model where two nu-
cleotides become/remain bounded, although such a situa-
tion might not occur experimentally. These situations are:
i) single complementary bases in the middle of two non-
complementary (or non-bound) ssDNA sequences; ii) sin-
gle complementary base positioned at the end of a non-
complementary (or not-bound) sequence; iii) single com-
plementary base, where the complement is on the same
ssDNA, at distance 0 or 1; iv) single complementary base
when the opposite neighbors of the pair are bound but not
complementary, i.e., are bound but not to each-other
7
The default value of the kinetic rate constant K
max
imple-
menting an instantaneous reaction is set to 10
5
SIMULTECH 2021 - 11th International Conference on Simulation and Modeling Methodologies, Technologies and Applications
290
Table 1: The list of 12 rules governing the dynamics of the DNA self-assembly process, and the default values of their
associated kinetic rate constants. The kinetic rate constants are scaled according to the rate of the intra-complex toehold
binding of complementary segments, i.e., Rule 4, which is normalized to value 1.
Rule# Description of rule action kinetic
param.
(k.p.)
default
val. of
k.p.
Rule 1 toehold binding of compl. seq. (3-to-7 bases) of 2 un-connected ssDNA k
1
0.001
Rule 2 binding of s.c.b.
1
of two immediately-connected ssDNA segments
2
k
2
300
Rule 3 binding of s.c.b. of two closely-connected (1-off neighbor connection) ssDNA
segments
2
k
3
30
Rule 4 toehold binding of compl. seq. (3-to-7 bases) of two connected ssDNA k
4
1
Rule 5 un-binding of s.c.b. positioned at a split
3
k
5
30
Rule 6 un-binding of s.c.b. positioned in the middle of a compl. sequence k
6
0.1
Rule 7 un-binding of s.c.b. where at least one ssDNA ends on that position k
7
50
Rule 8 very rapid (instantaneous) un-binding between s.c.b. in abnormal conditions
4
k
8
K
max
5
Rule 9 random unbinding of a pair of bound nucleotides k
9
0.01
Rule
10
rapid un-binding of a size-2 comp. seq. in the middle of non-compl./non-bound
seq.
k
10
300
Rule
11
rapid un-binding of a size-3 comp. seq. in the middle of non-compl./non-bound
seq.
k
11
100
Rule
12
rapid un-binding of a size-4 comp. seq. in the middle of non-compl./non-bound
seq.
k
12
10
sequences of an experiment to be the output multi-
ssDNA structures generated during previous simula-
tions. Moreover, the newly generated input file can be
saved as a new species file, using the ”Save as custom
species file” check-box, or it will become (by default)
the current test tube to be simulated.
Figure 2: The Create Test Tube environment for creat-
ing/editing test tube (.species) files.
The Run Experiment feature, see Fig. 3 is used to
numerically simulate the DNA assembly process from
a given input .species file, or (by default) from the cur-
rent test tube. The VDNA-Lab software uses NFsim
simulator on background to run experiments. NFsim
needs several parameters to run and produce output
(.dump) files for a certain simulation. The user can
enter numerical values for the Start time, Number of
dumps and Sim. end (i.e., end model-time for the sim-
ulation); the default values for these entries are 0, 10,
and 1, respectively. The output, i.e., .dump, files are
saved within a newly created folder, named after the
name of the input file and the time of the simulation;
the location of this folder is introduced by the user
using the Save simulation output directory tab.
The Run Experiment feature has also several Ad-
vanced options. By selecting this tab, a new win-
dow opens giving the user advanced choices regard-
ing three modeling aspects, each contained within a
delimited area of the window:
First, the user can select between using the default
BNGL model file
8
, describing the entire model’s en-
tities and all their interaction rules, or a separate user-
specified .bngl model file, possibly describing a user-
altered variant of this model; such possible model-
variant modifications are suggested with respect to
the custom temperature-controlled annealing proto-
cols detailed below.
8
The default BNGL model file is called master VDNA-
LAB BNGL model.bngl, and it is placed within the folder
... \system files\
˜
reference files of the VDNA-Lab instal-
lation folder
VDNA-Lab: A Computational Simulation Platform for DNA Multi-strand Dynamics
291
Figure 3: The Run simulation environment.
The second advanced feature offers the possibility
to modify the kinetic rate constants of the 12 reaction
rules indicated in Table 1, which are governing the
entire assembly (and dis-assembly) dynamics.
Finally, the third advanced feature allows the
user to define an annealing protocol for the simu-
lation/experiment, during which the temperature of
the system, and subsequently part of the reaction
rates, are modified several times mid simulation.
Namely, the user can specify the start and end tem-
perature of the system, the (constant) temperature
decrease/increase per annealing step, as well as the
length of time each annealing step will last. By de-
fault, without the annealing mechanism, the system
is assumed at a (normalized) temperature Temp =
1. When the annealing mechanism is activated, the
kinetic rate constants of reaction rules 5–12, from
within Table 1, are multiplied by the Temp values,
while the kinetic rate constants of reaction rules 1–
4 remain unmodified. If the user wants to create new
Temp-dependable kinetic rate constants for any of the
above rules, then he can modify the default BNGL
model file
8
within the ”begin functions ... end func-
tions” block, and then select the option of using this
custom .bngl model file.
The Visualize Test Tube window, see Fig. 4, can
be used by the user to generate a browser-based text-
enhanced visualization of the content of a simulation
output (.dump) file, see e.g. the visualization from
Fig. 5 associated to the running example discussed in
the next section. Within the visualization, each com-
plex, composed from one or several bound ssDNA,
is displayed separately, with each ssDNA within the
complex on a separate line. Per each ssDNA we rep-
Figure 4: The Visualized Test Tube environment.
resent the nucleotide sequence, either on the 5’ to 3’
order, or on the reverse order. The first ssDNA of
a complex is always displayed on the 5’ to 3’ or-
der, while subsequent ssDNAs within the complex
have the reversed order of the first complemented se-
quence. Moreover, if some particular nucleotide is
bound to another, by hovering the mouse over this
nucleotide its complement will highlight too. Also,
consecutive segments of bound nucleotides will have
a similar coloring. Due to these features, the user can
easily see the active domains of the complex, i.e., se-
quences of consecutive nucleotides bound to comple-
mentary contiguous sequences.
The Advanced option gives user further ability to
obtain filtered results, highlights, and also get stats
about the selected filtered complexes. Further details
made available by the visualization feature will be de-
scribed below, when we provide a running example of
our software on several simple input instances.
3.1 Running Examples
In this section, we present two running examples for
our computational simulation platform, one designed
to capture a DNA strand-displacement reaction, and
the other following the assembly of DAE-E DNA-
tiles (Jiang and et al., 2017). In both cases, VDNA-
Lab takes as input populations of isolated (or partially
bounded) oriented ssDNAs, and simulates their bind-
ing and dissociation reactions. The kinetic rate con-
stants used within these examples are the software de-
fault values, which means that the rates are normal-
ized and thus the current model-time is different from
what one might observe in a real wet-lab experiment.
SIMULTECH 2021 - 11th International Conference on Simulation and Modeling Methodologies, Technologies and Applications
292
In our first test case, for the first simulation, we
introduced 20 copies ot the ssDNAs:
ACGAT T GCT and CCAGATCGT
We simulated the assembly for 0.1 second (model-
time), and we visualize the content of the test-tube,
see Fig. 5 a) for a fraction of this visualization
9
.
On top of the visualization window, we display
some basic statistics providing the content of the test-
tube, namely the average, minimum and maximum
number of ssDNAs per complex. For example, for
our test-tube visualization in Fig. 5 a) we have that
the average number of ssDNAs per complex is 1.33,
while the minimum and maximum numbers of ss-
DNA within a complex are 1 and 2, respectively.
Within a visualization window, each distinct con-
figuration corresponding to a complex is visualized
only once. However, within a given test-tube (i.e.,
configuration) there could be many identical copies
of the same complex. In such a case, the amount of
identical copies are summed up and mentioned ac-
cordingly on top of each complex, as well as the num-
ber of ssDNAs that are included in that complex. For
example, regarding the first visualized complex from
Fig. 5 a) we can say that within that test-tube there are
9 complexes having the exactly same configuration.
Single ssDNA within a complex can be identified as
a chain of nucleotides placed on consecutive position.
Multiple ssDNAs within a complex appear each on
a separate line. The nucleotides that are bound to
a complementary pair appear in color, whereas non-
complemented nucleotides are displayed in gray. Mi-
nor color shift can be noticed within nucleotides that
are consecutively complemented or that are closely
located in an ssDNA.
Continuing our example, after the initial assem-
bly, using the Import from Species function from the
Create Test Tube window we add to the current con-
figuration 20 copies of the new ssDNA
AGCAATCGT
The simulation is restarted for another 0.5 seconds,
after which the output is visualized in Fig. 5 b)
10
.
As it can be observed from Fig. 5 b), one can ob-
serve both the situations were a strand displacement is
in process, for those complexes containing 3 ssDNA
molecules, and situations in which one of the initially
bounded ssDNA molecules has been completely dis-
placed, and only the two fully complementary strands
are either partial or completely bound.
9
The entire content of the test tube and thus of the visual-
ization window requires more space, and thus was omitted
10
As before, due to space limitation only a fraction of the
visualization window is displayed
Figure 5: Fraction of simulation output after 0.1s and sub-
sequent 0.5s after addition of new ssDNA.
Figure 6: Generated output of a DNA self-assembly system
simulation and the desired layout of a DAE-E DNA-tile.
In a completely different experiment we have cre-
ated a test-tube containing 20 copies of each of the 5
ssDNA molecules forming a DAE-E DNA-tile (Jiang
and et al., 2017). We have simulated their assembly
for 1 second and then visualized the output. In Fig. 6
we display one of the visualized complexes consist-
ing of 5 ssDNA bound together into a successfully as-
sembled DAE-E tile, as well as the desired schematic
representation of such a tile.
4 CONCLUSIONS
In this paper, we have introduced VDNA-
Lab (Spencer et al., 2021), a computational modeling
framework for DNA multi-strand dynamics using the
BioNetGen Language (BNGL) formalism (Harris,
2015) and the NFsim computational platform (Sned-
don and et al., 2011). The ssDNAs are provided as
input while the system simulates their binding and
dissociation reactions. The model implementation
is done with the help of 12 binding and un-binding
local interaction rules. Each of these rules has its
own, user-adjustable, kinetic rate constant and each
of them is implemented through one or several rule-
based reactions. Tracking and reporting the global
mapping of the components within a heterogeneous
complex is done with the help of a VDNA-Lab visu-
alization subroutine. The content of the dynamical
system is unloaded at various time-points within the
VDNA-Lab: A Computational Simulation Platform for DNA Multi-strand Dynamics
293
simulation, and it is re-assembled for visualization
and further numerical analysis. A simple 2D graphi-
cal representation of the assembled complexes is also
generated, where one can track the ssDNAs within
the complex as well as all the binding interactions.
Different from other computational modeling
frameworks for DNA strand assembly, we can modify
some of the systems parameters, such as the temper-
ature of the system, during the system simulation.
Thus, we can model also an entire annealing process
for the formation of a DNA structure. The underlining
computational modeling engine used by VDNA-Lab,
namely NFsim, allows also for other parameters to be
adjusted mid-simulation: e.g., we can define specific
binding/un-binding kinetic rates for each different
length-3 subsequence. Currently, VDNA-Lab does
not have these features implemented.
ACKNOWLEDGEMENTS
This work was partially supported by the Academy of
Finland, project 311371/2017, and by the Romanian
National Authority for Scientific Research and Inno-
vation, PED grant 2391.
REFERENCES
Ariga, K. and et al. (2008). Challenges and breakthroughs
in recent research on self-assembly. Science and Tech-
nology of Advanced Materials, 9(1):014109.
Blinov, M. L. and et al. (2004). BioNetGen: software for
rule-based modeling of signal transduction based on
the interactions of molecular domains. Bioinformat-
ics, 20(17):3289–3291.
Danos, V. and et al. (2009). Rule-based modelling and
model perturbation. In Transactions on Computa-
tional Systems Biology XI, pages 116–137. Springer.
Faeder, J. R., Blinov, M. L., and Hlavacek, W. S. (2009).
Rule-Based Modeling of Biochemical Systems with
BioNetGen, pages 113–167. Humana Press, NJ.
Gautam., V., Long., S., and Orponen., P. (2020). Ruledsd:
A rule-based modelling and simulation tool for dna
strand displacement systems. In Proceedings of the
13th International Joint Conference on Biomedical
Engineering Systems and Technologies - Volume 3:
BIOINFORMATICS,, pages 158–167.
Glotzer, S. C., Solomon, M. J., and Kotov, N. A. (2004).
Self-assembly: From nanoscale to microscale col-
loids. AIChE Journal, 50(12):2978–2985.
Harris, L. A. e. a. (2015). BioNetGen 2.2: Advances
in Rule-Based Modeling. arXiv e-prints, page
arXiv:1507.03572.
Jiang, S. and et al. (2017). Understanding the elemen-
tary steps in dna tile-based self-assembly. ACS Nano,
11(9):9370–9381.
Kamerlin, S. C. L. and Warshel, A. (2011). Multiscale
modeling of biological functions. Phys. Chem. Chem.
Phys., 13:10401–10411.
Krishnan, Y. and Simmel, F. C. (2011). Nucleic acid based
molecular devices. Angewandte Chemie International
Edition, 50(14):3124–3156.
Kuzyk, A., Laitinen, K. T., and T
¨
orm
¨
a, P. (2009). DNA
origami as a nanoscale template for protein assembly.
Nanotechnology, 20(23):235305.
Lakin, M. R. and et al. (2011). Visual dsd: a design
and analysis tool for dna strand displacement systems.
Bioinformatics, 27:3211–3213.
Li, J. and et al. (2011). Self-assembled multivalent dna
nanostructures for noninvasive intracellular delivery
of immunostimulatory cpg oligonucleotides. ACS
Nano, 5(11):8783–9.
Lund, K. and et al. (2010). Molecular robots guided by
prescriptive landscapes. Nature, 465(7295):206–209.
Mohammed, A., Czeizler, E., and Czeizler, E. (2017).
Computational modelling of the kinetic tile assembly
model using a rule-based approach. Theoretical Com-
puter Science, 701:203 215. At the intersection of
computer science with biology, chemistry and physics
- In Memory of Solomon Marcus.
Patitz, M. J. (2011). Simulation of self-assembly in the
abstract tile assembly model with ISU TAS. CoRR,
abs/1101.5151.
Poppleton, E. and et al. (2020). Design, optimization and
analysis of large DNA and RNA nanostruc through in-
teractive visualization, editing and molecular simula-
tion. Nucleic Acids Research, 48(12):e72–e72.
Seeman, N. (2003). Dna in a material world. Nature Cell
Biology, 421(6921):427–431.
Sneddon, M. and et al. (2011). Efficient modeling, simula-
tion and coarse-graining of biological complexity with
nfsim. Nat Methods, 8(2):177–183.
Spencer, F., Sanwal, U., and Czeizler, E. (2021). VDNA-
Lab. Available at https://github.com/Frankie-Spencer/
/virtual dna lab , version 1.0.
Sweeney, B. and et al. (2008). Exploring the parameter
space of complex self-assembly through virus capsid
models. Biophysical Journal, 94(3):772 – 783.
Thomas, M. and Schwartz, R. (2017). Quantitative compu-
tational models of molecular self-assembly in systems
biology. Physical Biology, 14(3):035003.
Yang, J. and Hlavacek, W. S. (2011). The efficiency of
reactant site sampling in network-free simulation of
rule-based models for biochemical systems. Physical
Biology, 8(5):055009.
SIMULTECH 2021 - 11th International Conference on Simulation and Modeling Methodologies, Technologies and Applications
294