Cargando…

Accurate Classification of Protein Subcellular Localization from High-Throughput Microscopy Images Using Deep Learning

High-throughput microscopy of many single cells generates high-dimensional data that are far from straightforward to analyze. One important problem is automatically detecting the cellular compartment where a fluorescently-tagged protein resides, a task relatively simple for an experienced human, but...

Descripción completa

Detalles Bibliográficos
Autores principales: Pärnamaa, Tanel, Parts, Leopold
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Genetics Society of America 2017
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5427497/
https://www.ncbi.nlm.nih.gov/pubmed/28391243
http://dx.doi.org/10.1534/g3.116.033654
_version_ 1783235637698101248
author Pärnamaa, Tanel
Parts, Leopold
author_facet Pärnamaa, Tanel
Parts, Leopold
author_sort Pärnamaa, Tanel
collection PubMed
description High-throughput microscopy of many single cells generates high-dimensional data that are far from straightforward to analyze. One important problem is automatically detecting the cellular compartment where a fluorescently-tagged protein resides, a task relatively simple for an experienced human, but difficult to automate on a computer. Here, we train an 11-layer neural network on data from mapping thousands of yeast proteins, achieving per cell localization classification accuracy of 91%, and per protein accuracy of 99% on held-out images. We confirm that low-level network features correspond to basic image characteristics, while deeper layers separate localization classes. Using this network as a feature calculator, we train standard classifiers that assign proteins to previously unseen compartments after observing only a small number of training examples. Our results are the most accurate subcellular localization classifications to date, and demonstrate the usefulness of deep learning for high-throughput microscopy.
format Online
Article
Text
id pubmed-5427497
institution National Center for Biotechnology Information
language English
publishDate 2017
publisher Genetics Society of America
record_format MEDLINE/PubMed
spelling pubmed-54274972017-05-12 Accurate Classification of Protein Subcellular Localization from High-Throughput Microscopy Images Using Deep Learning Pärnamaa, Tanel Parts, Leopold G3 (Bethesda) Investigations High-throughput microscopy of many single cells generates high-dimensional data that are far from straightforward to analyze. One important problem is automatically detecting the cellular compartment where a fluorescently-tagged protein resides, a task relatively simple for an experienced human, but difficult to automate on a computer. Here, we train an 11-layer neural network on data from mapping thousands of yeast proteins, achieving per cell localization classification accuracy of 91%, and per protein accuracy of 99% on held-out images. We confirm that low-level network features correspond to basic image characteristics, while deeper layers separate localization classes. Using this network as a feature calculator, we train standard classifiers that assign proteins to previously unseen compartments after observing only a small number of training examples. Our results are the most accurate subcellular localization classifications to date, and demonstrate the usefulness of deep learning for high-throughput microscopy. Genetics Society of America 2017-04-08 /pmc/articles/PMC5427497/ /pubmed/28391243 http://dx.doi.org/10.1534/g3.116.033654 Text en Copyright © 2017 Parnamaa and Parts http://creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
spellingShingle Investigations
Pärnamaa, Tanel
Parts, Leopold
Accurate Classification of Protein Subcellular Localization from High-Throughput Microscopy Images Using Deep Learning
title Accurate Classification of Protein Subcellular Localization from High-Throughput Microscopy Images Using Deep Learning
title_full Accurate Classification of Protein Subcellular Localization from High-Throughput Microscopy Images Using Deep Learning
title_fullStr Accurate Classification of Protein Subcellular Localization from High-Throughput Microscopy Images Using Deep Learning
title_full_unstemmed Accurate Classification of Protein Subcellular Localization from High-Throughput Microscopy Images Using Deep Learning
title_short Accurate Classification of Protein Subcellular Localization from High-Throughput Microscopy Images Using Deep Learning
title_sort accurate classification of protein subcellular localization from high-throughput microscopy images using deep learning
topic Investigations
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5427497/
https://www.ncbi.nlm.nih.gov/pubmed/28391243
http://dx.doi.org/10.1534/g3.116.033654
work_keys_str_mv AT parnamaatanel accurateclassificationofproteinsubcellularlocalizationfromhighthroughputmicroscopyimagesusingdeeplearning
AT partsleopold accurateclassificationofproteinsubcellularlocalizationfromhighthroughputmicroscopyimagesusingdeeplearning