رکورد قبلیرکورد بعدی

" Using Machine Learning to Predict and Analyze Biological Networks "


Document Type : Latin Dissertation
Language of Document : English
Record Number : 149686
Doc. No : ET21478
Main Entry : Manuel Johannes Middendorf
Title Proper : Using Machine Learning to Predict and Analyze Biological Networks
Note : This document is digital این مدرک بصورت الکترونیکی می باشد
Abstract : Recently, large amounts of experimental data for complex biological systems havebecome available. We here use modern tools from machine learning to build data-driven, predictive models. We first show how to infer network evolution mechanismsfor the Drosophila melanogaster protein-protein interaction network by learning aclassifier of network growth models. Our results show that Drosophila's networkmost closely obeys a duplication-mutation-complementation evolution principle.We then use information-theoretic principles to build an algorithm for mod-ule discovery in networks and give a precise mathematical definition of networkmodularity. Our algorithm outperforms other module discovery algorithms used inthe literature in correctly identifying modules for noisy data and in estimating theoptimal number of existing modules.We finally present a global model for transcriptional regulation predictive ofgenetic regulatory response in Saccharomyces cerevisiae. Our model learns putativetested for theQ1 PC1 bus cardBoth these projects mere sofixare des elopment efforts tonards contributing to dlfferentaspects of Roboucs and lZ1echatronics projects m the Controls and Roboucs Group..
Subject : Electericl tess
: برق
electronic file name : TL44648.pdf
Title and statement of responsibility and : Using Machine Learning to Predict and Analyze Biological Networks [Thesis]
آدرس ثابت

پیوستها
عنوان :
نام فایل :
نوع عام محتوا :
نوع ماده :
فرمت :
سایز :
عرض :
طول :
TL44648.pdf
TL44648.pdf
پایان نامه لاتین
متن
application/octet-stream
7.44 MB
85
85
نظرسنجی