Clustering Algorithms
| Hi-index | 0.00 |
With the explosion of biological data, manual data processing is no longer an option. Reliable automated software systems are urgently needed, especially in biomedical fields. In this paper, we present an automated scoring system for population SNP (single nucleotide polymorphisms) scores. This system uses our linear scanning density analysis algorithm to determine numbers of clusters and applies a maximum likelihood statistical algorithm to arrange scores according to similarity in pattern. The results can be displayed as a two-dimensional graph where the majority of scores are assigned to appropriate clusters in comparison to homozygous scores, designated as homozygous A, heterozygous AB, or homozygous B. We demonstrated that our integrated approach was able to achieve low cost and flexibility in the analysis of several population score data sets. Over 92.7% of population SNP scores can be clustered and the subsequent scores can be categorized into segregation groups. Out of 982 data set obtained from 96-well or 384-well plates, an 87.8% accuracy rate was achieved in assigning scores.