Dissertation
Ant colony optimization based inverse folding of mono- and bistable RNA macromolecules
Personen-Informationen ausblenden
Erschienen in
Bibliographische Angaben
DOI:
10.6094/UNIFR/11095
URN:
urn:nbn:de:bsz:25-freidok-110956
Sprache:
englisch
Informatik, Information und Wissen, allgemeine Werke
Erscheinungsjahr: 2016
Abstract
-
englisch
Since the discovery of structural conformations of DNA in the middle of the 20th century not only technologies that can elucidate structures of biologically relevant molecules have become more sophisticated, but the understanding of biological processes on the molecular level in general has grown tremendously in the last 60 years of research. In this same time, the early and dogmatic statement, according to which proteins are the only entities in molecular biological perception that can perform and provide necessary biological, e.g. enzymatic, functions within organisms, has undergone major revision. Of course, proteins do still perform the functions, which have been annotated, but in addition to the level of control of the proteins, specific RNA molecules, namely the non-coding RNAs, have been accounted to the executing functional level so far exclusive to proteins. As in the case of the proteins, a functional RNA receives its specific function from a biologically active structure conformation, which strongly correlates with the respective RNA sequence. A comparatively large number of sequences can fold into a similar structural RNA conformation. However, small perturbations as for example point mutations or sequestering of parts of the RNA sequence through other interacting entities can be key for the disruption of the functional structure of the RNA. Alongside with the exploration of new RNA functionalities, RNA based technologies have been derived from single RNA based functionalities and corresponding mechanisms. Their analytical and creative potential in combination with hereof derived computer programs, e.g. predicting structures from RNA sequences or vice versa, predicting RNA sequences from structures, extend the classical biological approach beyond its investigative origin by adding a progressive engineering spirit to the former purely research character.
In this dissertation, a computational RNA design tool and its application performance are presented. The tool is conceptually based on a relatively long heritage of tools, which can solve the 'RNA inverse fold' problem: Given a structure (mostly secondary RNA structure), the programs pursuit different strategies to produce a sequence, which can fold into the specified structure input. Classically a single structure was given as input. With the presented tool and its several capabilities of solving different levels of structural complexity based on RNA secondary structures, not only a new way of solving the problem with the heuristic approach of the ant-colony optimization technique was introduced. Furthermore, new constraints such as the regulation of a very precise GC content of the solution sequence has been given major concern in the concept as well as new structural constraint possibilities of pseudoknots and bistable RNA entities. The new introduced features are benchmarked and tested on structure complexity specific data sets, which have been gained from online data bases and corresponding literature efforts. Also comparative representations with other state of the art computer programs are given.
Beschreibung
-
Bemerkung
The presented approach is an adaptation of the ant colony optimization prinziple to the RNA inverse folding problem. As a core part, he document contains the formalistic concept and the benchmark of the implementation.
Dateien
Download
(3.54 MB)
Lizenz
Dissertation Robert Kleinkauf
Dissertation_RK_2016.pdf
SHA256 checksum: aac91ef0718738dfc3dc5e8711ba0cc402af9e4377dec5238255f752fb64bae4
Beschreibung der Forschungsdaten
Relationen
Laden...
- referenziert
Laden...
Korrekturanfrage
Vielen Dank für Ihre Korrekturanfrage. Wir werden uns um Ihre Anfrage kümmern und uns ggf. über die angegebene E-Mail-Adresse bei Ihnen zurückmelden. Bitte haben Sie Verständnis dafür, dass die Korrektur unter Umständen einige Tage dauern kann.
Es ist ein Fehler aufgetreten. Bitte versuchen Sie es später noch einmal.