Amino acid variation in human proteome

master's thesis

• Title (croatian): Aminokiselinska raznolikost ljudskog proteoma
• Author: Kristijan Vuković
• Mentor: Tomica Hrenar (mentor)
• Mentor: Michael Inouye https://orcid.org/0000-0001-9413-6520 (mentor)
• Committee member: Tomica Hrenar (predsjednik povjerenstva)
• Committee member: Zlatko Mihalić (član povjerenstva)
• Committee member: Marko Močibob (član povjerenstva)
• Committee member: Tajana Preočanin (član povjerenstva)
• Granter: University of Zagreb; Faculty of Science; (Department of Chemistry); Zagreb
• Defense date and country: 2016-09-30, Croatia
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Chemistry

Abstract

Amino acids are important biological molecules that influence human health and disease
through their function as structural units of proteins. Recent advancements to the genome
sequencing technologies enabled an indirect, large-scale exploration of human proteomic
diversity by mapping of nucleotide sequences in protein coding genomic regions to their
corresponding amino acid sequences in proteome. In this thesis, data from the 1000
Genomes and several other sequencing projects was used for the construction of amino
acid substitution maps that explore their occurrence frequency and pathogenicity.
Biochemical structural classification of amino acids was identified as an important element
in predicting both of these, with most class transitions showing the inversely proportional
relationship between the two. Codon distribution of the underlying genetic code was
compared with substitution maps and proved insufficient to account for the observed
frequencies. Classification accuracy of pathogenic variants from SIFT and PolyPhen, two
bioinformatical tools used in the 1000 Genomes Project was assessed, and frequency
resolved substitution maps for 5 population groups defined in the project were constructed
and analyzed. Finally, Trp  Ser, a substitution that continuously showed high pathogenicity
signal was further explored through a series of molecular dynamics simulations.

Abstract (croatian)

Aminokiseline su važne biološke molekule koje kao monomerne jedinice sudjeluju u izgradnji proteina.
Nedavni razvoj tehnologija za sekvenciranje genoma omogućio je indirektno određivanje njihovih sekvenci
mapiranjem nukleotidnih sljedova protein-kodirajućih regija genoma na odgovarajuće aminokiselinske
sljedove proteoma. Takva metoda višestruko je brža i jeftinija od direktnog sekvenciranja proteina, a nove
tehnologije revolucionarizirale su primjenu genetskog koda u medicinskim i znanstvenim istraživanjima. U
ovom radu, podatci iz projekta “1000 Genomes” i nekoliko sličnih projekata iskorišteni su za konstrukciju mapa
aminokiselinskih supstitucija s obzirom na njihovu učestalost i patogenost. Biokemijska klasifikacija
aminokiselina pokazala se ključnom u objašnjavanju dobivenih rezultata, a tranzicije između strukturnih grupa
u ove dvije mape imale su značajnu obrnuto proporcionalnu korelaciju koja je u skladu s evolucijskim
selektivnim pritiscima. Distribucija kodona pojedinih aminokiselina pokazala se nedostatnom za objašnjavanje
dobivenih supstitucijskih mapa, a s obzirom na nukleotidne pozicije unutar kodona, uočena je povećana
patogenost aminokiselinskih supstitucija uzrokovanih promjenom 2. te smanjena patogenost supstitucija
uzrokovanih promjenom 3. nukleotida. Testirana su dva bioinformatička alata za klasifikaciju patogenosti
aminokiselinskih supstitucija korištena u projektu “1000 Genomes” te se alat PolyPhen pokazao nešto boljim
od SIFT-a u detekciji patogenih supstitucija. Također, konstruirane su i analizirane frekvencijski razriješene
supstitucijske mape 5 populacijskih grupa definiranih u tom projektu. Konačno, supstitucija Trp  Ser, koja je
u više analiza pokazala značajan signal patogenosti, detaljnije je strukturno analizirana kroz seriju molekulsko
dinamičkih simulacija. Simulirani su proteini u kojima je dotična supstitucija detektirana i to uz pouzdanu
klasifikaciju patogenosti.

• Keywords: 1000 Genomes Project; bioinformatics; human proteome; molecular dynamics; sequencing technologies; statistical hypothesis testing
• Keywords (croatian): projekt 1000 Genomes; bioinformatika; humani proteom; molekulska dinamika; sekvenciranje genoma; statističko testiranje hipoteza
• Language: english
• URN:NBN: urn:nbn:hr:217:538899
• Study programme: Title: Graduate university studies in chemistry; specializations in: Research programme; Course: Research programme; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra kemije (magistar/magistra kemije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access; Embargo expiration date: 2017-09-30
• Created on: 2017-03-14 17:03:10