Single cell transcriptome data analysis of human and mice beta cells from diabetic and non-diabetic donors.

Lavrenčič, Eva (avtor), Jakše, Jernej (mentor), Theis, Fabian (komentor)

Pancreatic β-cells, which are a highly plastic, heterogeneous group of cells, produce insulin and have a key role in glucose homeostasis. Because malfunctioning of β-cells causes diabetes, they represent a potential target for diabetes regenerative therapies. Since most diabetes studies are done on animal models that are not directly comparable to humans, the translation of the findings from animal models to humans is needed. Here, we integrated, annotated and jointly analysed five human pancreatic islet scRNA-seq datasets, focusing on the variability of more than 50.000 β-cells in health and dysfunction. We compared the human atlas to a similar previously developed resource of murine data. We found that batch effects were less effectively removed in human data than in mice, possibly due to additional, unexplained heterogeneity in humans. Differential expression analysis of β-cells from middle-aged type 2 diabetic and healthy samples identified 24 human and 2049 mice differentially expressed genes (DEGs), of which four (ALDOB, CPNE4, ITIH5, LAMB1) were differentially expressed in the same direction in both organisms. We found that disease states could explain a relatively higher proportion of variability in mice than in humans, which could explain the difference in the number of identified DEGs. Lastly, we defined ten human β-cell gene programmes (GPs) as groups of spatially covariable genes, some of which were shared with previously defined mice gene programmes. Our discoveries also corresponded to previous observations on human data, for example, the data showed heterogeneous activity of β-cell insulin production and subsequent regeneration-associated GPs. We evaluated how likely each of the gene programs resulted from true biological variability or batch effects.

• Jezik: Angleški jezik
• Ključne besede: beta-cell heterogeneity, diabetes, humans, mice, pancreatic islets, scRNA-seq data analysis, species comparison
• Vrsta gradiva: Magistrsko delo/naloga
• Tipologija: 2.09 - Magistrsko delo
• Organizacija: BF - Biotehniška fakulteta
• Leto izida: 2024
• PID: 20.500.12556/RUL-158270
• COBISS.SI-ID: 198113795
• Datum objave v RUL: 01.06.2024

LAVRENČIČ, Eva, 2024, Single cell transcriptome data analysis of human and mice beta cells from diabetic and non-diabetic donors. [na spletu]. Magistrsko delo. [Dostopano 18 april 2025]. Pridobljeno s: https://repozitorij.uni-lj.si/IzpisGradiva.php?lang=slv&id=158270

Sekundarni jezik

• Jezik: Slovenski jezik
• Naslov: Analiza transkriptomskih podatkov posameznih zdravih in diabetičnih beta celic človeka ter miši

Izvleček

Celice β trebušne slinavke, ki proizvajajo inzulin, imajo ključno vlogo pri uravnavanju metabolizma glukoze. Celice β so plastična in heterogena skupina celic. Nepravilno delovanje celic β povzroča diabetes, zato le-te predstavljajo potencialno tarčo za zdravljenje diabetesa. Ker večino raziskav o diabetesu opravimo na živalskih modelih, ki niso neposredno primerljivi s človekom, je za boljše razumevanje heterogenosti celic β potreben prenos znanja iz živalskega modela na človeka. V nalogi smo integrirali, določili tipe celic in skupaj analizirali človeške transkriptomske podatke posameznih celic pankreasnih otočkov iz petih podatkovnih setov, pri čemer smo se osredotočili na variabilnost več kot 50.000 zdravih in nepravilno delujočih celic β ter izvedli primerjavo s podobnim predhodno razvitim virom mišjih podatkov. Ugotovili smo, da je bila odstranitev tehničnih vplivov med integracijo slabša pri človeku kot pri miših. V celicah β iz donorjev z diabetesom tipa 2 in zdravih donorjev srednjih let smo odkrili 24 človeških in 2049 mišjih diferencialno izraženih genov, od katerih so bili štirje (ALDOB, CPNE4, ITIH5, LAMB1) regulirani v isto smer. Spremenljivka bolezen je pri miših pojasnila relativno višji delež variabilnosti, kar bi lahko pojasnilo razliko v številu diferencialno izraženih genov. Določili smo deset genskih programov človeških β-celic (GP) kot skupine prostorsko spremenljivih genov, med katerimi smo našli GP-je, podobne predhodno definiranim mišjim GP-jem. Preverili smo, koliko so na določanje GP-jev vplivali biološki ali tehnični učinki. Opazili smo heterogenost celic β, kjer stanju visoke proizvodnje inzulina sledi faza regeneracije, kar se ujema z izsledki predhodnih raziskav pri človeku.

• Ključne besede: analiza podatkov scRNA-seq, diabetes, heterogenost celic beta, ljudje, miši, otočki trebušne slinavke, primerjava vrst