A neural network classifier capable of recognizing the patterns of all major subcellular structures in fluorescence microscope images of HeLa cells.

MOTIVATION: Assessment of protein subcellular location is crucial to proteomics efforts since localization information provides a context for a protein's sequence, structure, and function. The work described below is the first to address the subcellular localization of proteins in a quantitative, comprehensive manner. RESULTS: Images for ten different subcellular patterns (including all major organelles) were collected using fluorescence microscopy. The patterns were described using a variety of numeric features, including Zernike moments, Haralick texture features, and a set of new features developed specifically for this purpose. To test the usefulness of these features, they were used to train a neural network classifier. The classifier was able to correctly recognize an average of 83% of previously unseen cells showing one of the ten patterns. The same classifier was then used to recognize previously unseen sets of homogeneously prepared cells with 98% accuracy. AVAILABILITY: Algorithms were implemented using the commercial products Matlab, S-Plus, and SAS, as well as some functions written in C. The scripts and source code generated for this work are available at http://murphylab.web.cmu.edu/software. CONTACT: murphy@cmu.edu

Multivariate analysis.

UNLABELLED: One of the goals of Current Protocols is to provide information from the very basic level to the very advanced. This unit is an excellent example of providing explanation and application for a complex area. Building upon several other units on data analysis, it provides a detailed explanation of how multivariate analytical techniques are applied to flow cytometry data. The unit deals with the covariance matrix, principal component analysis, and cluster analysis and provides sources for software as well as example data files. This material will be particularly valuable to the graduate student trying to obtain a good understanding of statistical methods in flow cytometry, as well as to the experienced investigator who uses these tools on a regular basis without really understanding how they operate. KEYWORDS: principal components analysis; cluster analysis; FCS file format One of the goals of Current Protocols is to provide information from the very basic level to the very advanced.

Towards a systematics for protein subcelluar location: quantitative description of protein localization patterns and automated analysis of fluorescence microscope images.

Determination of the functions of all expressed proteins represents one of the major upcoming challenges in computational molecular biology. Since subcellular location plays a crucial role in protein function, the availability of systems that can predict location from sequence or high-throughput systems that determine location experimentally will be essential to the full characterization of expressed proteins. The development of prediction systems is currently hindered by an absence of training data that adequately captures the complexity of protein localization patterns. What is needed is a systematics for the subcellular locations of proteins. This paper describes an approach to the quantitative description of protein localization patterns using numerical features and the use of these features to develop classifiers that can recognize all major subcellular structures in fluorescence microscope images. Such classifiers provide a valuable tool for experiments aimed at determining the subcellular distributions of all expressed proteins. The features also have application in automated interpretation of imaging experiments, such as the selection of representative images or the rigorous statistical comparison of protein distributions under different experimental conditions. A key conclusion is that, at least in certain cases, these automated approaches are better able to distinguish similar protein localization patterns than human observers.

Toward objective selection of representative microscope images.

Scientists wishing to communicate the essential characteristics of a pattern (such as an immunofluorescence distribution) currently must make a subjective choice of one or two images to publish. We therefore developed methods for objectively choosing a typical image from a set, with emphasis on images from cell biology. The methods involve calculation of numerical features to describe each image, calculation of similarity between images as a distance in feature space, and ranking of images by distance from the center of the feature distribution. Two types of features were explored, image texture measures and Zernike polynomial moments, and various distance measures were utilized. Criteria for evaluating methods for assigning typicality were proposed and applied to sets of images containing more than one pattern. The results indicate the importance of using distance measures that are insensitive to the presence of outliers. For collections of images of the distributions of a lysosomal protein, a Golgi protein, and nuclear DNA, the images chosen as most typical were in good agreement with the conventional understanding of organelle morphologies. The methods described here have been implemented in a web server (http://murphylab.web.cmu.edu/services/TyplC).

Automated recognition of patterns characteristic of subcellular structures in fluorescence microscopy images.

Methods for numerical description and subsequent classification of cellular protein localization patterns are described. Images representing the localization patterns of 4 proteins and DNA were obtained using fluorescence microscopy and divided into distinct training and test sets. The images were processed to remove out-of-focus and background fluorescence and 2 sets of numeric features were generated: Zernike moments and Haralick texture features. These feature sets were used as inputs to either a classification tree or a neural network. Classifier performance (the average percent of each type of image correctly classified) on previously unseen images ranged from 63% for a classification tree using Zernike moments to 88% for a backpropagation neural network using a combination of features from the 2 feature sets. These results demonstrate the feasibility of applying pattern recognition methods to subcellular localization patterns, enabling sets of previously unseen images from a single class to be classified with an expected accuracy greater than 99%. This will provide not only a new automated way to describe proteins, based on localization rather than sequence, but also has potential application in the automation of microscope functions and in the field of gene discovery.