Primjena konvolucijske neuronske mreže u brojanju i lokalizaciji proteina fotografiranih super-rezolucijskom mikroskopijom

master's thesis

• Title (english): Application of convolutional neural network for protein counting and localization photographed by super-resolution microscopy
• Author: Ivan Meić
• Mentor: Carlo Manzo (mentor)
• Mentor: Damjan Franjević (sumentor)
• Committee member: Damjan Franjević (predsjednik povjerenstva)
• Committee member: Sunčica Bosak (član povjerenstva)
• Committee member: Inga Marijanović (član povjerenstva)
• Committee member: Goran Klobučar (član povjerenstva)
• Granter: University of Zagreb; Faculty of Science; (Department of Biology); Zagreb
• Defense date and country: 2020-02-28, Croatia
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Biology

Abstract

Razvoj novih tehnika mikroskopije tijekom proteklih nekoliko godina doveo je do razvoja super-rezolucijske mikroskopije s povećanom rezolucijom u odnosu na konvencionalne tehnike mikroskopije. Grana super-rezolucijske mikroskopije, bazirana je na lokalizaciji jedne molekule i uključuje tehnike “Photoactivated localization microscopy” (PALM) i “Stochastic Optical Reconstruction Microscopy” (STORM), koje omogućuju znanstvenicima proučavanje staničnih procesa s povećanom rezolucijom. Mikroskopija lokalizacije jedne molekule oslanja se na prikupljanje seta slika gdje samo podskup optički razlučivih fluorofora emitira svjetlost. Analiza tih slika omogućuje rekonstrukciju visoko kvalitetnih super-rezolucijskih slika. Ovaj diplomski rad predlaže sljedeći korak u analizi super-rezolucijskih slika, baziran na analizi lokalizacije pojedinog proteina. Rad upotrebljava višestruko brojanje signala jednog proteina za predviđanje pozicije proteina. Predviđena lokalizacija nadilazi artefakte višestrukog brojanja kroz upotrebu neuralne mreže, s ciljem brojanja i lokalizacije protein. Brojanje proteina sa slika super-rezolucijske mikroskopije predstavlja problem zbog višestrukog brojanja fluorofora. Za razliku od ostalih radova koji pokušavaju riješiti ovaj problem, ovaj diplomski rad ne koristi modeliranje nego se oslanja na sposobnost neuralne mreže da ući iz podataka. Ovaj pristup pruža znanstvenicima brzo izvršavajući algoritam koji zahtjeva minimalno znanje o neuralnim mrežama i analizi slika. Glavni cilj ovog diplomskog rada je izraditi i izmjeriti koliko dobro suvremena neuralna mreža za obradu slika može lokalizirati i prebrojati proteine.

Abstract (english)

In recent years novel microscopy techniques have led to the development of super-resolution imaging with increased resolution compared to conventional microscope techniques. A branch of super-resolution microscopy, based on single-molecule localization that includes techniques such as Photoactivated localization microscopy (PALM) and Stochastic Optical Reconstruction Microscopy (STORM), enables researchers to study cellular processes with increased resolution. Single-molecule localization microscopy relies on collecting a set of images where only a subset of optically resolvable fluorophores is emitting light. The analysis of these images allows for reconstruction of high-quality super-resolution images. This graduation thesis proposes a further step in super-resolution image analysis, based on the analysis of individual protein localization. It proposes to use the multiple counts produced by a single protein to predict its position. Position is predicted beyond overcounting artifact through a neural network, in order to predict the protein number and position. Counting proteins from super-resolution image is a problem because of fluorophore overcounting. Unlike previous papers trying to solve this problem, this graduation thesis does not use any modelling and relies only on neural network's ability to learn from data. This approach provides researchers with a fast performing algorithm that requires minimal knowledge about neural networks and image analysis. The main object of this graduation thesis is to create and measure how well the state of the art image processing neural network can localize and count proteins.

• Keywords: super-rezolucijska mikroskopija; CNN; strojno učenje; brojanje i lokalizacija proteina
• Keywords (english): super-resolution microscopy; CNN; machine learning; protein counting and localization
• Language: croatian
• URN:NBN: urn:nbn:hr:217:020655
• Study programme: Title: Molecular Biology; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra molekularne biologije (magistar/magistra molekularne biologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access; Embargo expiration date: 2022-02-28
• Created on: 2020-07-13 07:04:41