Padma Natarajan

Adaptive double self-organizing maps for clustering gene expression profiles

This paper introduces a new model of self-organizing map (SOM) known as adaptive double self-organizing map (ADSOM). ADSOM has a flexible topology and performs data partitioning and cluster visualization simultaneously without requiring a priori knowledge about the number of clusters. It combines features of the popular SOM with two-dimensional position vectors, which serve as a visualization tool to detect the number of clusters present in the data. ADSOM updates its free parameters and allows convergence of its position vectors to a fairly consistent number of clusters provided its initial number of nodes is greater than the expected number of clusters. A novel index is introduced based on hierarchical clustering of the final locations of position vectors. The index allows automated detection of the number of clusters, thereby reducing human error that could be incurred from counting clusters visually. To test ADSOMs consistency in data partitioning, we examine the number of common profiles found in the clusters that were obtained by varying the initial number of nodes. This provides a confidence measure for the clusters formed by ADSOM and illustrates the effect of different initial number of nodes on data partitioning. The reliance of ADSOM in identifying number of clusters is demonstrated by applying it to publicly available yeast gene expression data.

Applications of fuzzy logic in genomics

Advances in techniques for high throughput data gathering such as microarrays and DNA sequencing machines have opened up new research avenues in genomics. Large-scale biological research such as genome projects are now producing enormous quantities of genomic data using these rapidly growing technologies. Transforming the massive data to useful biological knowledge is the present challenge. Different analysis tools are being developed in order to detect and understand the phenomena of gene regulation and physiological functions and assessing the quality of a genomic sequence. Fuzzy systems are suitable for uncertain or approximate reasoning when systems are difficult to describe with a mathematical model. They allow problem solving and decision making with incomplete or uncertain information. This unique feature makes them an ideal tool for analyzing complex genomic data. This paper presents application of fuzzy systems in (1) developing a confidence measure to assess the accuracy of bases called by a DNA basecalling algorithm, and (2) building a gene interaction model that identifies triplets of activators, repressors, and targets in gene expression data. It is shown that applying appropriate fuzzy conjunction and aggregation rule increases the resilience of the fuzzy gene interaction model to noise.

Computational Intelligence and its Applicationin Remote Sensing

Remote sensing observations provide a new global perspective of the Earth environment. Measurements from airborne and space borne sensor systems help scientists gain a better understanding of the complex interactions between the Earth’s atmosphere, oceans, ice regions and land surfaces, as well as human-induced change due to population growth and human activities. These remote sensing measurements are widely used in geographical, meteorological, and environmental studies. Technological advancements have resulted in an increase in the number of observation platforms and sensor capabilities (e.g., spectral and spatial resolution). This trend will continue and soon will produce an unprecedented volume of data. Information extracted from these datasets will support national research agendas and national applications that will exert an ever-increasing requirement for shorter processing times and greater data and algorithm accuracies. Hence, advanced mathematical techniques are needed to effectively analyze data generated from the rapidly growing remote sensing technology. For most geophysical retrieval algorithms, adding additional information to improve the measurement of in situ properties is not a simple task because of the nonlinear nature of the problem as well as computational difficulties. Moreover, most current mathematical techniques generally require a high level of scientific knowledge of the physical system to accurately analyze remotely sensed data. In contrast, computational intelligence (CI) techniques such as artificial neural networks, genetic algorithms, and fuzzy logic systems, provide the capability to better examine complex data without requiring detailed knowledge about the underlying physical system. For example, CI techniques have been used to accurately estimate bio-optical parameters in complex coastal aquatic environments from remotely sensed data by employing special features such as the ability to learn from data, adaptive behavior, handling of non-linear systems, flexibility towards the choice of inputs, and resilience against noise. For example, in satellite remote sensing of ocean color, most algorithms are based on regression (or empirical) models that use power and/or cubic polynomials to relate ratios of remotely sensed reflectance to bio-optical parameters such as chlorophyll

Ensemble neural network methods for satellite-derived estimation of chlorophyll /spl alpha/

In this paper, neural network-based methods incorporating ensemble learning techniques are presented that estimate chlorophyll /spl alpha/ (chl /spl alpha/) concentration in the coastal waters of the Gulf of Maine (GOM). A dataset was constructed consisting of in situ chl measurements from the GOM matched with satellite data from the sea-viewing wide-field-of-view sensor (SeaWiFS). These data were used to develop models using diverse neural network ensembles for estimation of chl /spl alpha/ concentration from satellite-retrieved ocean reflectances. Results indicate that the models are able to generalize across geographical and temporal variation, and are resilient to uncertainty such as that introduced by poor atmospheric correction, or radiance contributions from non-chl /spl alpha/ components in case 2 waters.

Knowledge based extraction of ridge lines from digital terrain elevation data

A fuzzy rule based approach for extracting ridge lines from digital terrain elevation data (DTED) and digital elevation model (DEM) is presented. The algorithm consists of three major steps: Fuzzy rule based classification of the elevation data, ridge pixel detection from the pixel class under consideration, and filtration of unwanted pixels. The algorithm was tested on different DTED level 1 files and digital elevation model (DEM) and works well for both.

Neural network-based light attenuation model for monitoring seagrass population in the Indian river lagoon

Seagrasses have been considered one of the most critical marine habitat types of coastal and estuarine ecosystems such as the Indian River Lagoon. They are an important part of biological productivity, nutrient cycling, habitat stabilization and species diversity and are the primary focus of restoration efforts in the Indian River Lagoon. The areal extent of seagrasses has declined within segments of the lagoon over the years. Light availability to seagrasses is a major criterion limiting their distribution. Decreased water clarity and resulting reduced light penetration have been cited as the major factors responsible for the decline in seagrasses in the lagoon. Hence, light is a critical factor for the survival of seagrass species. Light attenuation coefficient is an important parameter that indicates the light attenuated by the water column and can therefore be used as an indicator of seagrass vigor. A number of region-specific linear light attenuation models have been proposed in the literature. Though, in practice, linear light attenuation models have been commonly used, there is need for a flexible and robust model that incorporates the non-linearities present in coastal and estuarine environments. This paper presents a neural network based model to estimate light attenuation coefficient from water quality parameters and thereby indirectly monitor seagrass population in the Indian River Lagoon. The proposed neural network models were compared with linear regression models, step-wise linear regression models, model trees and support vector machines. The neural network models performed fairly better compared to the other models considered.

Adaptive double self-organizing map and its application in gene expression data

This paper presents a novel clustering technique known as adaptive double self-organizing map (ADSOM). ADSOM has a flexible topology and perform clustering and cluster visualization simultaneously, thereby requiring no a priori knowledge about the number of clusters. ADSOM combines features of the popular self-organizing map (SOM) with two-dimensional position vectors, which serve as a visualization tool to accurately determine the number of clusters present in the data. ADSOM updates its free parameters during training and it allows convergence of its position vectors to a fairly consistent number of clusters provided that its initial number of nodes is greater than the expected number of clusters. A novel index is introduced based on hierarchical clustering of the final locations of position vectors. The index allows automatic detection of the number of clusters, thereby reducing human error that could be incurred from counting cluster visually. The reliance of ADSOM in identifying the number of clusters is proven by applying it to publicly available yeast gene expression data.

Neural network based light attenuation model for monitoring seagrass health

Light availability to seagrasses is a major criterion limiting the distribution of seagrasses. Decreased water clarity and resulting reduced light penetration have been cited as major factors responsible for the decline in seagrasses. Light attenuation coefficient is an important parameter that indicates the light attenuated by the water column and can thereby be an indicator of seagrass health. Though, in practice, linear light attenuation models have been commonly used, there is a need for a more accurate model that can take into account the non-linearities present in coastal and estuarine environments. This paper presents neural network-based light attenuation models for monitoring the seagrass health in the Indian River Lagoon, FL. For performance evaluation, results of the developed neural network models are compared with linear regression models, model trees, and support vector machines.

Monitoring seagrass health using neural networks

Monitoring seagrass health gives vital clues about the estuarine water quality, which is crucial for the existence of many aquatic plants and animals. Photosynthetic efficiency is a measure of plant stress and can be used to monitor seagrass health. However, insitu measurements of photosynthetic efficiency are time consuming and expensive. In this paper, neural network-based models are developed to estimate photosynthetic efficiency of a seagrass species, Zostera capricorni, from spectral reflectance measurements. The proposed neural network-based approach can be extended for other seagrass species by combining an ensemble of neural networks with a classifier. After identifying the type of seagrass species using the classifier, the neural network model that corresponds to the identified species is used to estimate photosynthetic efficiency.

Neural network-based estimation of chlorophyll-a concentration in coastal waters

The estimation of chlorophyll-a is one of the key indices of monitoring the phytoplankton populations. In this paper, an approach for estimating chlorophyll-a concentration using a neural network model is prose. A dat set assembled form various sources during the SeaWiFS Bio-optical Algorithm Mini-Workshop containing coincident in-situ chlorophyll and remote sensing reflectance measurements is used to evaluate the efficacy of the proposed neural network model. The data comprises of 919 stations and has chlorophyll-a concentrations ranging between 0.019 and 32.79 (mu) g/l. There are approximately 20 observations form more turbid coastal waters. A feed-forward neural network model with 10 noes in the hidden layer has been constructed to estimate chlorophyll-a concentration. The remote sensing reflectances form five SeaWiFS wavelengths are used as inputs to our model. The network is trained using the Levenberg-Marquardt algorithm. A neural network model can deal with non-linear relationships more accurately. Neural networks can effectively include variables that tend to co-vary non- linearly relationships more accurately. Neural networks can effectively include variables that tend to co-vary non- linearly with the output variable. They are flexible towards the choice of inputs and are tolerant to noise and require no a priori knowledge about the effect of these parameters. This makes them an ideal candidate for estimating chlorophyll-a concentration in coastal waters, where the presence of suspended sediments, detritus, and dissolved organic matter creates an optically complex situation. By allowing the neural network model to include several optical parameters as additional inputs to account for the scattering and absorption phenomena the model can be extended to estimate chlorophyll-a concentration turbid coastal waters.