[Hyperspectral image classification based on 3-D gabor filter and support vector machines].

A three-dimensional Gabor filter was developed for classification of hyperspectral remote sensing image. This method is based on the characteristics of hyperspectral image and the principle of texture extraction with 2-D Gabor filters. Three-dimensional Gabor filter is able to filter all the bands of hyperspectral image simultaneously, capturing the specific responses in different scales, orientations, and spectral-dependent properties from enormous image information, which greatly reduces the time consumption in hyperspectral image texture extraction, and solve the overlay difficulties of filtered spectrums. Using the designed three-dimensional Gabor filters in different scales and orientations, Hyperion image which covers the typical area of Qi Lian Mountain was processed with full bands to get 26 Gabor texture features and the spatial differences of Gabor feature textures corresponding to each land types were analyzed. On the basis of automatic subspace separation, the dimensions of the hyperspectral image were reduced by band index (BI) method which provides different band combinations for classification in order to search for the optimal magnitude of dimension reduction. Adding three-dimensional Gabor texture features successively according to its discrimination to the given land types, supervised classification was carried out with the classifier support vector machines (SVM). It is shown that the method using three-dimensional Gabor texture features and BI band selection based on automatic subspace separation for hyperspectral image classification can not only reduce dimensions; but also improve the classification accuracy and efficiency of hyperspectral image.

[A method of object detection for remote sensing-imagery based on spectral space transformation].

Object detection is an intermediate link for remote sensing image processing, which is an important guarantee of remote sensing application and services aspects. In view of the characteristics of remotely sensed imagery in frequency domain, a novel object detection algorithm based on spectral space transformation was proposed in the present paper. Firstly, the Fourier transformation method was applied to transform the image in spatial domain into frequency domain. Secondly, the wedge-shaped sample and overlay analysis methods for frequency energy were used to decompose signal into different frequency spectrum zones, and the center frequency values of object's features were acquired as detection marks in frequency domain. Finally, object information was detected with the matched Gabor filters which have direction and frequency selectivity. The results indicate that the proposed algorithm here performs better and it has good detection capability in specific direction as well.

[Hyperspectral remote sensing estimation models for snow grain size].

Snow grain size is a key parameter not only to affect the energy budget of the global or local region but also characterizing the status of snow vapor transport and temperature gradient. It is significant to monitor and estimate the snow grain size in large area for global or local climate change and water resource management. Recently, remote sensing technology has become a useful tool for snow grain size monitoring and estimating. In the present paper, the estimate models were built based on simulating the snow surface spectral reflectance curve in visible-infrared region and the sensitive bands and snow indices for snow grain size were selected. These models help estimate snow grain size by hyperspectral remote sensing. Through validating with ground true data, the results show that these models have higher explorative accuracy using 1 030, 1 260 nm and normalized difference snow index (460 and 1 030 nm). In addition, the correlation slopes of estimated and observed valves are 1.37, 0.61 and 0.62, respectively. R2 are 0.82, 0.86 and 0.93 and RMSE are 55.65, 50.83 and 35.91 microm, respectively. The result can provide a scientific basis for snow grain size monitoring and estimating.

Global analysis of in vivo EGR1-binding sites in erythroleukemia cell using chromatin immunoprecipitation and massively parallel sequencing.

Early growth response gene 1 (EGR1) has been implicated in megakaryocyte differentiation induced by phorbol ester. But the molecule mechanism of EGR1 in this process has not been widely investigated. The identification of direct EGR1 target genes in a global scale is critical for our understanding of how EGR1 contributes to this process. In this study, we provide a global survey on the binding location of EGR1 in the K562 cells using chromatin immunoprecipitation and massively parallel sequencing. Over 14 000 highly confident in vivo EGR1 binding sites were identified in phorbol 12-myristate 13-acetate-treated K562 cells. More than 70% of these genomic sites associated with EGR1 binding were located around annotated gene regions. Molecular functional classification of 6138 putative EGR1 target genes showed that the transcription factor class (695 of 6138; 11%) is the largest significantly enriched one. The results showed that a high coverage of the genome and a high positive rate achieve were achieved. This whole genome study on the EGR1 targets may provide a better understanding of the EGR1 regulated genes and the downstream pathway in megakaryocyte differentiation.

[Study on the genotyping of single nucleotide polymorphisms for a large number of samples by three-dimensional polyacrylamide gel-based microarray method].

OBJECTIVE: To genotype single nucleotide polymorphisms (SNPs) in a large number of samples by applying three-dimensional polyacrylamide gel-based microarray. METHODS: The method relies on copolymerization of acrylamide-modified PCR products with acrylamide monomers and acryl-modified slides to prepare gel-based microarray. Then array is hybridized with a pair of specific probes and the two universal dual-color fluorescent detectors labeled with Cy3 or Cy5 respectively (Tag1 and Tag2). Electrophoresis is used in post-hybridization to remove the nonspecifically bound targets and mismatches. Finally, genotyping is based on the images captured through two-color fluorescent scanning. RESULTS: The 3-D gel-immobilization of nucleic acids has a high immobilization yield and good hybridization efficiency. As universal dual-color fluorescent detectors are used, it is not required that specific probes be labeled for all SNPs, therefore the expense for synthesis can be reduced considerably. Electrophoresis in post-hybridization can enhance the capability for discriminating a single nucleotide mismatch from the perfectly matched sequence and improve the signal-to-noise ratio significantly. CONCLUSION: The gel-based microarray is a rapid, simple and high-throughput method for SNPs genotyping and may be very competitive in the efficiency, fidelity and cost for constructing DNA microarrays, which will hold significant promise for applications in human DNA diagnostics.

Correlation of the SNPs of FGFR1, FGF10, FGF18 with nonsyndromic cleft lip with or without palate in Chinese population.

OBJECTIVE: To explore the relationship between the polymorphisms in gene FGFR1, FGF10, FGF18 and the nonsyndromic cleft lip with or without cleft palate (NS CLP) in Chinese population. METHODS: Genomic DNA was isolated from peripheral lymphocytes of 75 patients with NS CLP and their parents and 75 unimpaired healthy children. The polymorphisms in FGFR1 gene rs13317, p.E467K, p.M369I and p.S393S, FGF10 gene rs1448037 and FGF18 gene rs4043716 were detected by applying three-dimensional (3-D) polyacrylamide gel microarray technology. The data were performed using statistical analysis: the genotype frequency and allele frequency between patients with NSCL/P and control subjects were performed. Haplotype relative risk (HRR), family based association test (FBAT), and transmission disequilibrium test (TDT) in nuclear family were performed. RESULTS: There were no polymorphism in FGFR1 gene p. E467K, p. M369I and p.S393S site, the corresponding base was all G. The polymorphisms of rs13317 and rs1448037 were detected and their genotype frequency and allele frequency showed no significant difference between 75 patients with NSCL/P and 75 normal children. TDT, HRR and FBAT were also no significant differences. The genotype frequency of gene FGF18 rs4043716 showed significant difference, but allele frequency were no significant difference. TDT, HRR and FBAT were also no significant difference. CONCLUSION: Our studies suggest an association between gene FGF18 rs4043716 and the NS CLP in Chinese population, and no association among gene FGFR1 rs13317, p. E467K, p. M369I, p. S393S and gene FGF10 rs1448037.

An improved gel-based DNA microarray method for detecting single nucleotide mismatch.

3-D polyacrylamide gel-based DNA microarray platforms provide a high capacity for nucleic acids immobilization and a solution-mimicking environment for hybridization. However, several technological bottlenecks still remain in these platforms, such as difficult microarray preparation and high fluorescent background, which limit their application. In this study, two new approaches have been developed to improve the convenience in microarray preparation and to reduce the background after hybridization. To control the polymerization process, solutions containing acrylamide-modified oligonucleotide, acrylamide, glycerol and ammonium persulfate are spotted onto a functionalized glass slide, and then the slide is transferred to a vacuum chamber with TEMED, so that TEMED is vaporized and diffused into the spots to induce polymerization. By applying an electric field across a hybridized microarray to remove the nonspecifically bound labeled targets, this approach can solve the problem of high fluorescent background of the gel-based microarray after hybridization. Experimental results show that our immobilization method can be used to construct high quality microarrays and exhibits good reproducibility. Moreover, the polymerization is not affected by PCR medium, so that PCR products can be used for microarray construction without being treated by commercial purification cartridges. Electrophoresis can improve the signal-to-noise significantly and has the ability to differentiate single nucleotide variation between two homozygotes and a heterozygote. Our results demonstrated that this is a reliable novel method for high-throughput mutation analysis and disease diagnosis.