diplomsko delo
Jurij Nastran (Author), Miha Moškon (Mentor), Žiga Pušnik (Co-mentor)

Abstract

V okviru diplomske naloge predstavimo postopek analize robusnosti modelov sintetičnih bioloških gradnikov. Postopek analize demonstiramo na primeru modela biološkega procesorja. Robustnost ocenimo na podlagi volumna prostora dopustnih parametrov v visokodimenzionalnem prostoru. Ta volumen izračunamo z uporabo genetskih algoritmov, metode glavnih komponent, nazadnje pa z metodo Monte Carlo. V diplomski nalogi demonstriramo pomembnost učinkovitega reduciranja omejujočega prostora možnih Monte Carlo vzorčenj za pridobitev natančnejšega rezultata z manj ponovitvami. Namesto hiperkvadra predlagamo uporabo hiperelipsoidov. Implementiramo metodo naključnega enakomernega vzorčenja znotraj danega hiperelipsoida. Primerjamo tudi dva pristopa za računanje omejujočega hiperelipsoida s čim manjšim volumnom. Z uporabo teh metod demonstriramo prednosti in slabosti uporabe hiperelipsoidov kot alternativi hiperkvadra pri izbiri omejujočih teles. Pri nekaterih množicah dopustnih parametrov vodi manjši volumen omejujočega hiperelipsoida k redukciji potrebnih vzorčenj Monte Carlo za več kot magnitudo.

Keywords

analiza robustnosti;vzorčenje Monte Carlo;hiperkvader;hiperelipsoid;volumen;sintezna biologija;modeliranje;simulacija;računalništvo;računalništvo in informatika;univerzitetni študij;diplomske naloge;

Data

Language: Slovenian
Year of publishing:
Typology: 2.11 - Undergraduate Thesis
Organization: UL FRI - Faculty of Computer and Information Science
Publisher: [J. Nastran]
UDC: 004.8:575.112(043.2)
COBISS: 31471363 Link will open in a new window
Views: 1071
Downloads: 223
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: Robustness assessment of the biological processor using hyperellipsoids
Secondary abstract: As part of this thesis, we present extentions of the current methods for the analysis of the robustness of synthetic biological systems. We demonstrate the new methods on a proposed programmable biological processor. The robustness is calculated using the volume of the body of viable parameters in high dimensional space. This volume is estimated using a genetic algorithm, principal component analysis and lastly Monte Carlo method. We demonstrate the importance the minimum enclosing body has on the efficiency of our method. Thus we focus our attention on improving the fitting of the enclosing body around the viable points. Instead of hypercuboids we demonstrate the advantages of the usage of hyperellipsoids as a viable alternative. For this purpose, a method for random uniform sampling inside of hyperellipsoid is implemented. Two competing methods of finding the minimum volume enclosing ellipsoids are also compared. We demonstrate the usefulness of the hyperellipsoid methods over the hypercuboid approach using the data provided by the biological processor model. In some cases, the ellipsoid approach is shown to reduce the number of required samplings by almost a magnitude.
Secondary keywords: robustness analysis;Monte Carlo sampling;hypercuboid;hyperellipsoid;volume;synthetic biology;modelling;simulation;computer science;computer and information science;diploma thesis;
Type (COBISS): Bachelor thesis/paper
Study programme: 1000468
Embargo end date (OpenAIRE): 1970-01-01
Thesis comment: Univ. v Ljubljani, Fak. za računalništvo in informatiko
Pages: 47 str.
ID: 12037028