My plan is to develop a tool that predicts mutations in user input genome. I have collected some set of genes and their corresponding sequence. The nucleotide sequence positions of genes are collected from hg19 reference genome. So the idea is when user inputs the sequence data it maps with collected sequence data and detects variation. But my doubts are,
1. How reliable the chromosome location given in hg19 reference genome? Does the given position contains exons, introns and other regions? or it specify only coding region?
2. Basically, what are the algorithms works behind predicting mutations in a genome?
3. Can single change in nucleotide on mapping be considered as mutation? Is there any frequency measures to confirm the mutation?
Answers are expectded in details
Thank you
What you describe sounds like methods for genome comparisons used to construct the Conservation tracks in the UCSC Genome Browser (details described at the bottom of this page: https://genome.ucsc.edu/cgi-bin/hgTrackUi?g=cons100way ; software used includes Multiz, Lastz, PipMaker etc.)
If input genome is unassembled (in the form of NGS reads), the procedure is called "variant detection" and there is plenty of info on software - "bowtie2 and samtools mpileup" come to mind. The results are commonly stored as VCF files.
Databases of single nucleotide polymorphisms (SNP) are also relevant to what you addressed in your question (https://www.ncbi.nlm.nih.gov/projects/SNP/).
@Matej Lexa what you have answered is fine. But expected answer is not that. I need to develop a tool offline that predicts the mutation. Please check 2nd and 3rd part of the question and let me know your answer
- Badges
- Science topic