Matching Items (2)
Learning Sparse Representations for Fruit-Fly Gene Expression Pattern Image Annotation and Retrieval
Description
Background
Fruit fly embryogenesis is one of the best understood animal development systems, and the spatiotemporal gene expression dynamics in this process are captured by digital images. Analysis of these high-throughput images will provide novel insights into the functions, interactions, and networks of animal genes governing development. To facilitate comparative analysis, web-based interfaces have been developed to conduct image retrieval based on body part keywords and images. Currently, the keyword annotation of spatiotemporal gene expression patterns is conducted manually. However, this manual practice does not scale with the continuously expanding collection of images. In addition, existing image retrieval systems based on the expression patterns may be made more accurate using keywords.
Results
In this article, we adapt advanced data mining and computer vision techniques to address the key challenges in annotating and retrieving fruit fly gene expression pattern images. To boost the performance of image annotation and retrieval, we propose representations integrating spatial information and sparse features, overcoming the limitations of prior schemes.
Conclusions
We perform systematic experimental studies to evaluate the proposed schemes in comparison with current methods. Experimental results indicate that the integration of spatial information and sparse features lead to consistent performance improvement in image annotation, while for the task of retrieval, sparse features alone yields better results.
Fruit fly embryogenesis is one of the best understood animal development systems, and the spatiotemporal gene expression dynamics in this process are captured by digital images. Analysis of these high-throughput images will provide novel insights into the functions, interactions, and networks of animal genes governing development. To facilitate comparative analysis, web-based interfaces have been developed to conduct image retrieval based on body part keywords and images. Currently, the keyword annotation of spatiotemporal gene expression patterns is conducted manually. However, this manual practice does not scale with the continuously expanding collection of images. In addition, existing image retrieval systems based on the expression patterns may be made more accurate using keywords.
Results
In this article, we adapt advanced data mining and computer vision techniques to address the key challenges in annotating and retrieving fruit fly gene expression pattern images. To boost the performance of image annotation and retrieval, we propose representations integrating spatial information and sparse features, overcoming the limitations of prior schemes.
Conclusions
We perform systematic experimental studies to evaluate the proposed schemes in comparison with current methods. Experimental results indicate that the integration of spatial information and sparse features lead to consistent performance improvement in image annotation, while for the task of retrieval, sparse features alone yields better results.
ContributorsYuan, Lei (Author) / Woodard, Alexander (Author) / Ji, Shuiwang (Author) / Jiang, Yuan (Author) / Zhou, Zhi-Hua (Author) / Kumar, Sudhir (Author) / Ye, Jieping (Author) / Biodesign Institute (Contributor) / Center for Evolution and Medicine (Contributor) / Ira A. Fulton School of Engineering (Contributor) / College of Liberal Arts and Sciences (Contributor) / School of Life Sciences (Contributor)
Created2012-05-23
Description
Large quantities of sodic and alkaline bauxite residue are produced globally as a by-product from alumina refineries. Ecological stoichiometry of key elements [nitrogen (N) and phosphorus (P)] plays a critical role in establishing vegetation cover in bauxite residue sand (BRS). Here we examined how changes in soil chemical properties over time in rehabilitated sodic and alkaline BRS affected leaf N to P stoichiometry of native species used for rehabilitation. Both Ca and soil pH influenced the shifts in leaf N:P ratios of the study species as supported by consistently significant positive relationships (P < 0.001) between these soil indices and leaf N:P ratios. Shifts from N to P limitation were evident for N-fixing species, while N limitation was consistently experienced by non-N-fixing plant species. In older rehabilitated BRS embankments, soil and plant indices (Ca, Na, pH, EC, ESP and leaf N:P ratios) tended to align with those of the natural ecosystem, suggesting improved rehabilitation performance. These findings highlight that leaf N:P stoichiometry can effectively provide a meaningful assessment on understanding nutrient limitation and productivity of native species used for vegetating highly sodic and alkaline BRS, and is a crucial indicator for assessing ecological rehabilitation performance.
ContributorsGoloran, Johnvie B. (Author) / Chen, Chengrong (Author) / Phillips, Ian R. (Author) / Elser, James (Author) / College of Liberal Arts and Sciences (Contributor) / School of Life Sciences (Contributor)
Created2015-10-07