This project aims to develop a software that can handle the base calling of Tm-shifted melting curve SNP genotyping methods. A suprevised algorithm is implemented so that the base calling coefficients can be pre-trained for making calls of the new samples.'
Melting curve SNP (McSNP) is a relatively cheap and handy genotyping technique as it is an extension of the RT-PCR technique. It make use of the difference in melting temperature of PCR products for different alleles.
A supervised algorithm is more adequate for McSNP calling a clinicl setting than unsupervised algorithms.
A biallelic SNP usually has three types, AA, AB and BB. A base caling algorithm is to identify the true type among the three different types. Since the three types are in an order, the ordinal regression is adequate for the training.
Intended Use of the Software
Users and clinicians can conduct the base calling using the default settings directly. Or alternatively, they can carry out the coefficients training themselves. A graphical user interface is provided for the ease of data manipulation.
The software is implemented in the Java language. We will constantly update the new versions with enhanced functionality.
Download McSNP software version 1.0 beta: http://www.bioinformatics.org/mcsnp/wiki/java-code.rar
Demo Input file
Download demonstration file here: http://www.bioinformatics.org/mcsnp/wiki/web_demo.txt.