DeepSort: deep convolutional networks for sorting haploid maize seeds.

BACKGROUND: Maize is a leading crop in the modern agricultural industry that accounts for more than 40% grain production worldwide. THe double haploid technique that uses fewer breeding generations for generating a maize line has accelerated the pace of development of superior commercial seed varieties and has been transforming the agricultural industry. In this technique the chromosomes of the haploid seeds are doubled and taken forward in the process while the diploids marked for elimination. Traditionally, selective visual expression of a molecular marker within the embryo region of a maize seed has been used to manually discriminate diploids from haploids. Large scale production of inbred maize lines within the agricultural industry would benefit from the development of computer vision methods for this discriminatory task. However the variability in the phenotypic expression of the molecular marker system and the heterogeneity arising out of the maize genotypes and image acquisition have been an enduring challenge towards such efforts. RESULTS: In this work, we propose a novel application of a deep convolutional network (DeepSort) for the sorting of haploid seeds in these realistic settings. Our proposed approach outperforms existing state-of-the-art machine learning classifiers that uses features based on color, texture and morphology. We demonstrate the network derives features that can discriminate the embryo regions using the activations of the neurons in the convolutional layers. Our experiments with different architectures show that the performance decreases with the decrease in the depth of the layers. CONCLUSION: Our proposed method DeepSort based on the convolutional network is robust to the variation in the phenotypic expression, shape of the corn seeds, and the embryo pose with respect to the camera. In the era of modern digital agriculture, deep learning and convolutional networks will continue to play an important role in advancing research and product development within the agricultural industry.

Systems biology-based identification of Mycobacterium tuberculosis persistence genes in mouse lungs.

UNLABELLED: Identifying Mycobacterium tuberculosis persistence genes is important for developing novel drugs to shorten the duration of tuberculosis (TB) treatment. We developed computational algorithms that predict M. tuberculosis genes required for long-term survival in mouse lungs. As the input, we used high-throughput M. tuberculosis mutant library screen data, mycobacterial global transcriptional profiles in mice and macrophages, and functional interaction networks. We selected 57 unique, genetically defined mutants (18 previously tested and 39 untested) to assess the predictive power of this approach in the murine model of TB infection. We observed a 6-fold enrichment in the predicted set of M. tuberculosis genes required for persistence in mouse lungs relative to randomly selected mutant pools. Our results also allowed us to reclassify several genes as required for M. tuberculosis persistence in vivo. Finally, the new results implicated additional high-priority candidate genes for testing. Experimental validation of computational predictions demonstrates the power of this systems biology approach for elucidating M. tuberculosis persistence genes. IMPORTANCE: Mycobacterium tuberculosis, the causative agent of tuberculosis (TB), has a genetic repertoire that permits it to persist in the face of host immune responses. Identification of such persistence genes could reveal novel drug targets and elucidate mechanisms by which the organism eludes the immune system and resists drugs. Genetic screens have identified a total of 31 persistence genes, but to date only 15% of the ~4,000 M. tuberculosis genes have been tested experimentally. In this paper, as an alternative to brute force experimental screens, we describe computational methods that predict new persistence genes by combining known examples with growing databases of biological networks. Experimental testing demonstrated that these predictions are highly accurate, validating the computational approach and providing new information about M. tuberculosis persistence in host tissues. Using the new experimental results as additional input highlights additional genes for testing. Our approach can be extended to other data types and target organisms to characterize host-pathogen interactions relevant to this and other infectious diseases.

Predicting functional associations from metabolism using bi-partite network algorithms.

BACKGROUND: Metabolic reconstructions contain detailed information about metabolic enzymes and their reactants and products. These networks can be used to infer functional associations between metabolic enzymes. Many methods are based on the number of metabolites shared by two enzymes, or the shortest path between two enzymes. Metabolite sharing can miss associations between non-consecutive enzymes in a serial pathway, and shortest-path algorithms are sensitive to high-degree metabolites such as water and ATP that create connections between enzymes with little functional similarity. RESULTS: We present new, fast methods to infer functional associations in metabolic networks. A local method, the degree-corrected Poisson score, is based only on the metabolites shared by two enzymes, but uses the known metabolite degree distribution. A global method, based on graph diffusion kernels, predicts associations between enzymes that do not share metabolites. Both methods are robust to high-degree metabolites. They out-perform previous methods in predicting shared Gene Ontology (GO) annotations and in predicting experimentally observed synthetic lethal genetic interactions. Including cellular compartment information improves GO annotation predictions but degrades synthetic lethal interaction prediction. These new methods perform nearly as well as computationally demanding methods based on flux balance analysis. CONCLUSIONS: We present fast, accurate methods to predict functional associations from metabolic networks. Biological significance is demonstrated by identifying enzymes whose strong metabolic correlations are missed by conventional annotations in GO, most often enzymes involved in transport vs. synthesis of the same metabolite or other enzyme pairs that share a metabolite but are separated by conventional pathway boundaries. More generally, the methods described here may be valuable for analyzing other types of networks with long-tailed degree distributions and high-degree hubs.

Metabolic flux correlations, genetic interactions, and disease.

Many diseases are caused by failures of metabolic enzymes. These enzymes exist in the context of networks defined by the static topology of enzyme-metabolite interactions and by the reaction fluxes that are feasible at steady state. We use the local topology and the flux correlations to identify how failures in the metabolic network may lead to disease. First, using yeast as a model, we show that flux correlations are a powerful predictor of pairwise mutations that lead to cell death -- more powerful, in fact, than computational models that directly estimate the effects of mutations on cell fitness. These flux correlations, which can exist between enzymes far-separated in the metabolic network, add information to the structural correlations evident from shared metabolites. Second, we show that flux correlations in human align with similarities in Mendelian phenotypes ascribed to known genes. These methods will be useful in predicting genetic interactions in model organisms and understanding the combinatorial effects of genetic variations in humans.

Adaptive epileptic seizure prediction system.

Current epileptic seizure "prediction" algorithms are generally based on the knowledge of seizure occurring time and analyze the electroencephalogram (EEG) recordings retrospectively. It is then obvious that, although these analyses provide evidence of brain activity changes prior to epileptic seizures, they cannot be applied to develop implantable devices for diagnostic and therapeutic purposes. In this paper, we describe an adaptive procedure to prospectively analyze continuous, long-term EEG recordings when only the occurring time of the first seizure is known. The algorithm is based on the convergence and divergence of short-term maximum Lyapunov exponents (STLmax) among critical electrode sites selected adaptively. A warning of an impending seizure is then issued. Global optimization techniques are applied for selecting the critical groups of electrode sites. The adaptive seizure prediction algorithm (ASPA) was tested in continuous 0.76 to 5.84 days intracranial EEG recordings from a group of five patients with refractory temporal lobe epilepsy. A fixed parameter setting applied to all cases predicted 82% of seizures with a false prediction rate of 0.16/h. Seizure warnings occurred an average of 71.7 min before ictal onset. Similar results were produced by dividing the available EEG recordings into half training and testing portions. Optimizing the parameters for individual patients improved sensitivity (84% overall) and reduced false prediction rate (0.12/h overall). These results indicate that ASPA can be applied to implantable devices for diagnostic and therapeutic purposes.

Measuring information flow in nonlinear systems--a modeling approach in the state space.

Directional information flow between coupled nonlinear systems is of practical interest in many areas like bioengineering, chemistry, physics and electrical engineering. Due to the high complexity and nonlinearity of the coupled chaotic systems, linear modeling approaches may fail to capture the proper dynamics and thus the proper directional information flow. This necessitates novel approaches to analyze signals derived from such systems. This paper proposes a novel approach for detecting such directional information flows between the subsystems involved. The dependability of the method is illustrated using coupled chaotic oscillators in various coupling configurations.