alternate assembly

[litali]

Hi all,
I wanted to ask what is alternate assembly in blast and how I should treat it. I mean that I find structural differences when I am doing a mapping project, but then, when I check it in blast, I find that it firs perfectly alternate assembly, so it is not a real variation? how can I avoid this situation?
THANKS!!!

[Jon_Keats]

The alternate assembly or often assemblies are just that alternate build of your given genome from different individuals. The simple solution is to first pick the reference genome you want to use then in blast settings select to blast just this reference so you don't get hits from all the possible assemblies. Unfortunately, Watson and Venter are not identical twins so even their builds are not identical...go figure

[litali]

alternate assembly

Thank you. Yes, I understand that there might be assemblies from different people and I can set the settings in blast not to see it, but the question is not whether I see it or not but whether the structural difference I found when mapping to the reference genome is a mutation, and if I understand correctly, if there is an alternate assembly like this, this is not a real mutation?

[Jon_Keats]

Hi all,
I wanted to ask what is alternate assembly in blast and how I should treat it. I mean that I find structural differences when I am doing a mapping project, but then, when I check it in blast, I find that it firs perfectly alternate assembly, so it is not a real variation? how can I avoid this situation?
THANKS!!!

True versus false variation is dependent on your research question. If you are looking for novel disease causing structural variations you can use blast as you are doing to filter out things in the other assemblies or use some of the various databases to exclude known variants in other datasets. But if you are looking of misrepresented variants you are talking more about annotation and counting versus filtering. But the short answer is if the variant is in another assembly it is not a real/novel variant. So you should filter your reference genome variant list against a list of known variants which are available in a number of databases