Title :
DCN variations for autism diagnosis
Author :
Alsagabi, Majid I. ; Tewfik, Ahmed H.
Author_Institution :
Dept. of Electr. & Comput. Eng., Univ. of Minnesota, Minneapolis, MN, USA
Abstract :
Autism is correlated with DNA copy number variations (DCV). We used a statistical approach to analyze 75Mb of DNA samples of 71 autistic children and 71 typically developing children to prove the existence of significant differences between the two groups in their copy number burden. We applied unsupervised Fuzzy k-means clustering analysis first locally to identify the most distinctive locations, then globally to synthesize a collection of CNVs that have a strong power of segregating AU group from TD. We defined a collection of 13 CNVs with a total length < 100kb that classifies 121 individuals out of 142 correctly.
Keywords :
DNA; fuzzy logic; genetics; medical computing; medical disorders; molecular biophysics; paediatrics; patient diagnosis; pattern classification; statistical analysis; DCN variations; DCV; DNA copy number variations; autism diagnosis; autistic children; copy number burden; developing children; statistical approach; unsupervised fuzzy k-means clustering analysis; Autism; Bioinformatics; Communication system control; DNA; Genomics; Gold; Hidden Markov models; Process control; Signal processing; Testing;
Conference_Titel :
Communications, Control and Signal Processing (ISCCSP), 2010 4th International Symposium on
Conference_Location :
Limassol
Print_ISBN :
978-1-4244-6285-8
DOI :
10.1109/ISCCSP.2010.5463374