DNA Barcode classification using General Regression Neural Network with different distance models
| Autor: | Alfonso Urso, Antonino Fiannaca, Riccardo Rizzo, Massimo La Rosa |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2015 |
| Předmět: | |
| Zdroj: | Mathematical Models in Biology, pp. 119–132. Berlin: Springer International Publishing AG, 2015 info:cnr-pdr/source/autori:M La Rosa, A. Fiannaca, R. Rizzo, A. Urso/titolo:DNA Barcode classification using General Regression Neural Network with different distance models/titolo_volume:Mathematical Models in Biology/curatori_volume:/editore: /anno:2015 Mathematical Models in Biology ISBN: 9783319234960 |
| Popis: | The “cythosome c oxidase subunits 1” (COI) gene is used for identification of species, and it is one of the so-called DNA barcode genes. Identification of species, even using DNA barcoding can be difficult if the biological examples are degraded. Spectral representation of sequences and the General Regression Neural Network (GRNN) can give some interesting results in these difficult cases. The GRNN is based on the distance between the memorized examples of sequence and the input unknown sequence, both represented using a vector space spectral representation. In this paper we will analyse the effectiveness of different distance models in the GRNN implementation and will compare the obtained results in the classification of full length sequences and degraded samples. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|