Applying K-means Clustering Research Articles

Cluster analysis tests the importance of myogenic gene expression during myofiber hypertrophy in humans

We applied K-means cluster analysis to test the hypothesis that muscle-specific factors known to modulate protein synthesis and satellite cell activity would be differentially expressed during progressive resistance training (PRT, 16 wk) in 66 human subjects experiencing extreme, modest, and failed myofiber hypertrophy. Muscle mRNA expression of IGF-I isoform Ea (IGF-IEa), mechanogrowth factor (MGF, IGF-IEc), myogenin, and MyoD were assessed in muscle biopsies collected at baseline (T1) and 24 h after the first (T2) and last (T3) loading bouts from previously untrained subjects clustered as extreme responders (Xtr, n=17), modest responders (Mod, n=32), and nonresponders (Non, n=17) based on mean myofiber hypertrophy. Myofiber growth averaged 2,475 microm2 in Xtr, 1,111 microm2 in Mod, and -16 microm2 in Non. Main training effects revealed increases in all transcripts (46-83%, P<0.005). For the entire cohort, IGF-IEa, MGF, and myogenin mRNAs were upregulated by T2 (P<0.05), while MyoD did not increase significantly until T3 (P<0.001). Within clusters, MGF and myogenin upregulation was robust in Xtr (126% and 65%) and Mod (73% and 41%) vs. no changes in Non. While significant in all clusters by T3, IGF-IEa increased most in Xtr (105%) and least in Non (44%). Although MyoD expression increased overall, no changes within clusters were detected. We reveal for the first time that MGF and myogenin transcripts are differentially expressed in subjects experiencing varying degrees of PRT-mediated myofiber hypertrophy. The data strongly suggest the load-mediated induction of these genes may initiate important actions necessary to promote myofiber growth during PRT, while the role of MyoD is less clear.

Fungal Detection in Wheat Using Near-Infrared Hyperspectral Imaging

Different species of fungi infect grain in the field and storage facilities. Contamination by fungi in grain is detected and quantified by traditional methods, such as microbial incubation and microscopic detection, which are subjective, labor intensive, and time consuming. An accurate and timely detection technique for fungal growth in grain is needed to prevent grain from spoiling and to reduce quality loss. In this study, the potential of near-infrared hyperspectral imaging to detect fungal infection in wheat was investigated. Wheat kernels infected with storage fungi, namely Penicillium spp., Aspergillus glaucus, and Aspergillus niger, were scanned using a hyperspectral imaging system, and a total of 20 image slices at evenly spaced wavelengths between 1000 to 1600 nm were acquired to form a hypercube. A multivariate image analysis (MIA) technique based on principal component analysis (PCA) was used to reduce the dimensionality of the image hypercubes. Two-class and four-class classification models were developed by applying k-means clustering and discriminant (linear, quadratic, and Mahalanobis) analyses. Two-class discriminant classification models gave maximum classification accuracy of 100%, and on average 97.8% infected kernels were correctly classified by the linear discriminant classifier. The four-class linear discriminant classifier correctly classified more than 95% of the kernels infected with Penicillium and 91.7% healthy kernels. However, the discriminant classifiers misclassified the kernels infected with A. niger and A. glaucus.

A Redefinition of Economic Regions in the U.S.

Since the 1950s the Bureau of Economic Analysis (BEA) has grouped the states into eight regions based primarily on cross-sectional similarities in their socioeconomic characteristics. This is the most frequently used grouping of states in the U.S. for economic analysis. Since several recent studies concentrate on similarities and differences in regional business cycles, this paper groups states into regions based not on a broad set of socioeconomic characteristics but on the similarities in their business cycles. The analysis makes use of a consistent set of coincident indexes estimated from a Stock and Watson-type model. We applied k-means cluster analysis to the cyclical components of these indexes to group the 48 contiguous states into eight regions with similar cycles. Having grouped the states into regions, we determine the relative strength of cohesion among the states in the various regions. Finally, we compare the regions defined in this paper with the BEA regions.

An Alternative Definition of Economic Regions in the United States Based on Similarities in State Business Cycles

Since the 1950s the Bureau of Economic Analysis (BEA) has grouped the states into eight regions based primarily on cross-sectional similarities in their socioeconomic characteristics. This paper groups states into regions based on the similarities in their business cycles. We applied k-means cluster analysis to the cyclical components of Stock-Watson-type indices estimated at the state level to group the 48 contiguous states into eight regions with similar cycles. We then compare the cohesion of the regions so defined with the cohesion of the BEA regions. Finally, we examine how that definition affects the results of some recent regional business cycle analysis.