Hi, Im learning about next gen. seq. So, with NGS you shear up dna into fragments and when the computer tries to align by matching overlapping regions isnt there a possibility of the fragments aligning in the wrong place? I'm assuming that the error rate would increase with the number of matching base pairs. Let's say your sequence is:
....ATG CAG NNN NNN AGT TAA CAG NNN NNN AGT TGC....
Isn't their a possibility of the bold region sequences being swaped? Sorry if this question is stupid, I'm just a newbie
Yes, there is usually a subset of reads that map in multiple locations and the mapping algorithms take care of this in different ways--based on the score calculated during the mapping process.
You will always have a probability of miss-assembly, nevertheless the algorithms are optimized to deal with this situations and are able to solve them in most cases. While conducting a NGS experiment (probably) you will have a relatively high coverage with high quality in each position, so your algorithm will use this information to solve most of the dubious positions. Also, while performing a de novo assembly, there are alignment criteria that should be taken into account, like minimum alignment identity and similarity between the reads that "neighbor" reads.
There is allways a possibilty of error. But most programs will tell you if a read was assigned to 1 or more places indicating problematic parts of your assembly. By selecting better software you can improve your assembly. Here is an article that tried to compare different methods of genome assembly using NGS http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0017915
I agree with the previous comments, of course there is always an assembly error rate, but as others already told you, mostly software are able to reduce it as much as possible. Then you have to inspect into your results and use software to remove contaminants, and so on.. I used SPAdes for my de novo assembly and it work very well, and you can read about comparison of assembler to choose the best that fit with your data.
https://insidedna.me/tutorials/view/benchmark-seven-popular-genome-assemblers
https://blogs.warwick.ac.uk/zzhou/entry/comparison_of_soapdenovo_1/
Here you can find some comparisons, and about the software, here the link to the article and the website to download it:
http://online.liebertpub.com/doi/abs/10.1089/cmb.2012.0021
http://bioinf.spbau.ru/en/spades
I hope it helps you .
Regards,
Domenico.
- Badges
- Science method
- Similar topics
- Biological Science
- Genetics
- Genetic Analysis
- genetic techniques
- Sequencing