Entering edit mode
23 months ago
yujen.chang
•
0
Hi there,
We have about 1000 in-house samples (PE 150bp 20M) and want to build our catalogue based on de novo assembly.
At begining, we intend to devide those samples into 10 chunks (namely about 100 samples per batch) to conduct the co-assembly. However, my mentor guided me to first group samples based on their reads (I have tried metacluster). But the groups are very uneven.
So this is my question: What is the most desirable way to build MAG catalogue ? I have referred to some paper, which tend to use co-assembly strategy instead grouping samples together first.
I am very confused about that, and need some experts to discuss over this.
Thank you in advance.
