Background: Fluorescence microscopy is widely used to determine the subcellular location of proteins. Efforts to determine location on a proteome-wide basis create a need for automated methods to analyze the resulting images. Over the past ten years, the feasibility of using machine learning methods to recognize all major subcellular location patterns has been convincingly demonstrated, using diverse feature sets and classifiers. On a well-studied data set of 2D HeLa single-cell images, the best performance to date, 91.5%, was obtained by including a set of multiresolution features. This demonstrates the value of multiresolution approaches to this important problem.

We propose a new active mask algorithm for the segmentation of fluorescence microscope images of punctate patterns. It combines the (a) flexibility offered by active-contour methods, (b) speed offered by multiresolution methods, (c) smoothing offered by multiscale methods, and (d) statistical modeling offered by region-growing methods into a fast and accurate segmentation tool. The framework moves from the idea of the "contour" to that of "inside and outside", or, masks, allowing for easy multidimensional segmentation. It adapts to the topology of the image through the use of multiple masks. The algorithm is almost invariant under initialization, allowing for random initialization, and uses a few easily tunable parameters. Experiments show that the active mask algorithm matches the ground truth well, and outperforms the algorithm widely used in fluorescence microscopy, seeded watershed, both qualitatively as well as quantitatively.

No abstract

We propose an algorithm for the classification of fluorescence microscopy images depicting the spatial distribution of proteins within the cell. The problem is at the forefront of the current trend in biology towards understanding the role and function of all proteins. The importance of protein subcellular location was pointed out by Murphy, whose group produced the first automated system for classification of images depicting these locations, based on diverse feature sets and combinations of classifiers. With the addition of the simplest multiresolution features, the same group obtained the highest reported accuracy of 91.5% for the denoised 2D HeLa data set. Here, we aim to improve upon that system by adding the true power of multiresolution-adaptivity. In the process, we build a system able to work with any feature sets and any classifiers, which we denote as a Generic Classification System (GCS). Our system consists of multiresolution (MR) decomposition in the front, followed by feature computation and classification in each subband, yielding local decisions. This is followed by the crucial step of combining all those local decisions into a global one, while at the same time ensuring that the resulting system does no worse than a no-decomposition one. On a nondenoised data set and a much smaller number of features (a combination of texture and Zernicke moment features) and a neural network classifier, we obtain a high accuracy of 89.8%, effectively proving that the space-frequency localized information in the subbands adds to the discriminative power of the system.

We present a novel active mask framework for the segmentation of fluorescence microscope images of cells, and in particular, for the segmentation of the Golgi body as well as cellvolume computation. We demonstrate that the algorithm is able to efficiently segment a stack of images and successfully assign multiple pieces of the Golgi body in a 2D image to the cell to which they belong. Further, we demonstrate that our algorithm is more accurate than manual segmentation of these images.

Abstract. Predicting the outcome of a game using players strength and weakness against the players of the opponent team by considering the statistics of a set of matches played by players helps captain and coaches to select the team and order the players. In this paper, we propose a supervised learning method using SVM model with linear, and nonlinear poly and RBF kernals to predict the outcome of the game against particular side by grouping the players at different levels in the order of play for both the teams. The comparison among different groups of players at same level gives the order of groups which contributes to winning probability. we also propose to develop a system which recommends a player for a specific role in a team by considering the past performances. we achieve this by finding the similar players by clustering all the players using k-means clustering and finding the five nearest players using k nearest neighbor (KNN) classifier. We calculate the ranking index for players using the game and players statistics extracted from a particular tournament. Experimental results demonstrate that, the n-dimensional data considered for modeling is not linearly separable. Hence the nonlinear SVM with RBF kernel outperforms from the linear and poly kernel. SVM with RFB kernel yields the accuracy of 75, precision of 83.5 and recall rate of 62.5. So we recommend the use of SVM with RBF kernel for game outcome prediction.

We propose an adaptive multiresolution (MR) approach for classification of fluorescence microscopy images of subcellular protein locations, providing biologically relevant information. These images have highly localized features both in space and frequency which naturally leads us to MR tools. Moreover, as the goal of the classification system is to distinguish between various protein classes, we aim for features adapted to individual proteins. These two requirements further lead us to adaptive MR tools. We start with a simple classification system consisting of Haralick texture feature computation followed by a maximum-likelihood classifier, and demonstrate that, by adding an MR block in front, we are able to raise the average classification accuracy by roughly 10%. We conclude that selecting features in MR subspaces allows us to custom-build discriminative feature sets for fluorescence microscopy images of protein subcellular location images.

The most common definition, provided by the WHO, for an adverse drug reaction (ADR) is "A response to a drug which is noxious and unintended and which occurs at doses normally used in man for prophylaxis, diagnosis, or therapy of disease, or the modification of physiological function or noxious and unintended responses to drugs, when they are administered in their normal recommended dosages." This study focuses on the prediction of ADRs caused by the drug-drug interaction (DDI) of two-drug combinations. Apart from contributing to various productive drug design strategies such as drug repurposing (Zhou et al., 2015), coadministered drugs can exhibit synergistic DDIs (Liu et al., 2017), which comprises a new ADR that may be associated with either of the drugs or the aggravation of an existing ADR. In this study, we have proposed an artificial neural network (ANN) that predicts this specific subclass of ADRs using transcriptomic data, compound chemical fingerprint, and GO ontologies.