diplomsko delo
Miha Svetelšek (Author), Tomaž Curk (Mentor)

Abstract

V diplomski nalogi smo modelirali povezavo med genotipom in fenotipom tridesetih vzorcev kvasovke S. cerevisiae. Na podlagi podatkov in predznanja smo določili mutacije posameznih nukleotidov in z njimi povezane gene, s katerimi je možno zgraditi dober model za napovedovanje fenotipa. Poleg določanja pomembnih mest v genomu (SNV-jev) nam zgrajeni model omogoča tudi določevanje pomembnih genotipov oziroma starševskega izvora, ki je povezan z opazovanim fenotipom. Vrednotenje modelov pokaže, da lahko z linearno regresijo zanesljivo napovedujemo fenotip. Fenotip relativno dobro napoveduje tudi model, ki je zgrajen le na podlagi podatkov o dveh izvornih starših in začetne populacije. Empirično smo določili povezavo med številom vzorcev, ki jih uporabimo za izgradnjo napovednih modelov, in napovedno napako modelov.

Keywords

bioinformatika;genotip;fenotip;posameznik;populacija;linearna regresija;logistična regresija;računalništvo;računalništvo in informatika;računalništvo in matematika;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: [M. Svetelšek]
UDC: 004.9:57(043.2)
COBISS: 10718036 Link will open in a new window
Views: 49
Downloads: 4
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: Predicting the phenotype from genotype data on individual and pooled segregants
Secondary abstract: We have modeled the relationship between genotype and phenotype using data on thirty yeast S. cerevisiae samples. Using prior knowledge, we have determined mutations of individual nucleotides and related genes with which it is possible to build a good prediction model for the phenotype. The constructed models allow us to determine the location of important mutations in the genome (SNVs), to rank samples based on phenotype, and to determine signi_cant genotypes or parental origin, which is connected to the observed phenotype. Evaluation of these models shows that the phenotype can be predicted very reliably with linear regression. The phenotype can be predicted relatively well from data on two starting parents and the _rst pool of segregants. We also show the relation between the number of samples used to build a predictive model and its predictive error.
Secondary keywords: bioinformatics;genotype;phenotype;individual segregant;pool of segregants;linear regression;logistic regression;computer science;computer and information science;computer science and mathematics;diploma;
File type: application/pdf
Type (COBISS): Undergraduate thesis
Thesis comment: Univerza v Ljubljani, Fakulteta za računalništvo in informatiko
Pages: 78 str.
ID: 24214993