Real-Time Protein Crystallization Image Acquisition and Classification System.

Real-Time Protein Crystallization Image Acquisition and Classification System.
复制标题

DOI:
10.1021/cg3016029
复制
发表时间:
2013-07-03
影响因子:
3.8
通讯作者:
Aygun RS
Aygun RS
中科院分区:
化学2区
文献类型:
--
作者:
Sigdel M;Pusey ML;Aygun RS

文献摘要

参考文献

被引文献

相似文献

在本文中,我们描述了一个独立的实时系统的蛋白质结晶图像采集和分类的目标,以协助结晶学家在评分结晶试验的设计和实施。内部组装的荧光显微镜系统用于图像采集。这些图像被分为三类,如非晶体,可能的铅,和晶体。图像分类包括两个主要步骤-图像特征提取和基于多层感知器(MLP)神经网络的分类应用。我们的特征提取涉及应用多种阈值技术、识别高强度区域(斑点)以及生成强度和斑点特征以获得每张图像的45维特征向量。为了降低丢失晶体的风险,我们引入了一个最大类集成分类器,它应用多个分类器,并选择最高的分数(或类)。我们在2250张图像上进行了实验,其中67%为非晶体图像,18%为可能的线索图像,15%为清晰的晶体图像,并使用10倍交叉验证测试了我们的结果。我们的实验结果表明,该方法是非常有效的(< 3秒处理和分类的图像),并具有较高的准确性。我们的系统只错过了1.2%的晶体(分类为非晶体),最有可能是由于低照明或失焦图像捕获,并具有88%的整体准确度。
In this paper, we describe the design and implementation of a stand-alone real-time system for protein crystallization image acquisition and classification with a goal to assist crystallographers in scoring crystallization trials. In-house assembled fluorescence microscopy system is built for image acquisition. The images are classified into three categories as non-crystals, likely leads, and crystals. Image classification consists of two main steps – image feature extraction and application of classification based on multilayer perceptron (MLP) neural networks. Our feature extraction involves applying multiple thresholding techniques, identifying high intensity regions (blobs), and generating intensity and blob features to obtain a 45-dimensional feature vector per image. To reduce the risk of missing crystals, we introduce a max-class ensemble classifier which applies multiple classifiers and chooses the highest score (or class). We performed our experiments on 2250 images consisting 67% non-crystal, 18% likely leads, and 15% clear crystal images and tested our results using 10-fold cross validation. Our results demonstrate that the method is very efficient (< 3 seconds to process and classify an image) and has comparatively high accuracy. Our system only misses 1.2% of the crystals (classified as non-crystals) most likely due to low illumination or out of focus image capture and has an overall accuracy of 88%.
DOI: 10.1007/s10969-009-9076-9
发表时间: 2010-03
期刊: Journal of structural and functional genomics
影响因子: --
作者:
Cumbaa, Christian A;Jurisica, Igor
通讯作者: Jurisica, Igor
DOI: 10.1107/s090744490602943x
发表时间: 2006-10-01
影响因子: 2.2
作者:
Berry, Ian M.;Dym, O.;Messerschmidt, Albrecht
通讯作者: Messerschmidt, Albrecht
DOI: 10.1107/s0907444905041648
发表时间: 2006-03-01
影响因子: 2.2
作者:
Pan, S;Shavit, G;Meldrum, D
通讯作者: Meldrum, D
DOI: 10.1107/s0907444902016840
发表时间: 2002-11-01
期刊: ACTA CRYSTALLOGRAPHICA SECTION D-BIOLOGICAL CRYSTALLOGRAPHY
影响因子: --
作者:
Spraggon, G;Lesley, SA;Priestle, JP
通讯作者: Priestle, JP
DOI: 10.1016/0022-0248(91)90878-9
发表时间: 1991-03-01
影响因子: 1.8
作者:
ZUK, WM;WARD, KB
通讯作者: WARD, KB