Machine Learning for Microbial Phenotype Prediction

This thesis presents a scalable, generic methodology for microbial phenotype prediction based on supervised machine learning, several models for biological and ecological traits of high relevance, and the deployment in metagenomic datasets. The results su

  • PDF / 3,716,635 Bytes
  • 116 Pages / 419.528 x 595.276 pts Page_size
  • 78 Downloads / 233 Views

DOWNLOAD

REPORT


Machine Learning for Microbial Phenotype Prediction

BestMasters

Springer awards „BestMasters“ to the best master’s theses which have been completed at renowned universities in Germany, Austria, and Switzerland. The studies received highest marks and were recommended for publication by supervisors. They address current issues from various fields of research in natural sciences, psychology, technology, and economics. The series addresses practitioners as well as scientists and, in particular, offers guidance for early stage researchers.

Roman Feldbauer

Machine Learning for Microbial Phenotype Prediction

Roman Feldbauer Wien, Österreich

BestMasters ISBN 978-3-658-14318-3 ISBN 978-3-658-14319-0 (eBook) DOI 10.1007/978-3-658-14319-0 Library of Congress Control Number: 2016940340 Springer Spektrum © Springer Fachmedien Wiesbaden 2016 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. Printed on acid-free paper This Springer Spektrum imprint is published by Springer Nature The registered company is Springer Fachmedien Wiesbaden GmbH

V

Kurzfassung Genomdatenbanken wachsen rasant. Moderne Metagenomikstudien führen zu einer großen Anzahl annähernd vollständiger Genomsequenzen nicht kultivierbarer mikrobieller Spezies. Diese Entwicklungen führen zur Notwendigkeit der Entwicklung automatisierter bioinformatischer Methoden für die Vorhersage mikrobieller Phänotypen, um die biologische und ökologische Interpretation der großen Datenmengen zu ermöglichen. In dieser Arbeit wird untersucht, wie komparative Genomik für diesen Zweck eingesetzt werden kann. Verschiedene bioinformatische Prototypen sowie Techniken des maschinellen Lernens werden verglichen. Im Fokus stehen dabei große Genomdatenbanken und inkomplette Genomsequenzen. Darüberhinaus werden notwendige Verbesserungen an der Software vorgenommen. Ein Programm wurde in der Evaluationsphase ausgewählt. Die Stabilität der Vorhersagen phänotypischer Charakteristika wurde im Lichte schnell wachsender Genomdatenbanken demonstriert