Hello!
I am struggling with a hairy problem and finally decided to look for help with knowledgeable fellows: I am trying to replicate a de novo assembly done by someone else.
We have contracted BGI to do WGS of a fish species (with estimated genome sizes of around 1,5 Gb) using HiSeq 2000, with 90 bases paired-end reads. They used a single short library of 500 bp fragments.
The aim was to characterize microsatellite loci in the resulting assembly.
We received the filtered short sequence data (16 Gbases) and a table with more than 80.000 different microsatellite loci, along with primers, 'left' (5') flank, 'right' (3') flank and sequence of the PCR product (which together span up to around 150-300 bp). We did not receive the associated scaffolds and contigs from where this data was obtained. We have empirically validated dozens of resulting loci so far.
The problem is I want to retrieve an assembly wich contains all the loci from the table. BGI sent us some of the alleged parameters (such as kmer=47), but we can't generate an assembly (using SOAPdenovo 2) that yield all (or even most) of the loci as described in BGI's table. I have not succeed to get more help from them on that issue.
Parameters used:
SOAPdenovo all -K 47 -F -R -D -d -M 1 -s lib.cfg
lib.cfg
avg_ins=500
reverse_seq=0
asm_flags=3
rank=1
pair_num_cutoff=3
map_len=32
So my question is: is it possible to independently obtain the very same assembly they originally generated?
What can I do to get closer to this in case we are not 100% sure on the original parameters.
Thanks in advance and best wishes
I am struggling with a hairy problem and finally decided to look for help with knowledgeable fellows: I am trying to replicate a de novo assembly done by someone else.
We have contracted BGI to do WGS of a fish species (with estimated genome sizes of around 1,5 Gb) using HiSeq 2000, with 90 bases paired-end reads. They used a single short library of 500 bp fragments.
The aim was to characterize microsatellite loci in the resulting assembly.
We received the filtered short sequence data (16 Gbases) and a table with more than 80.000 different microsatellite loci, along with primers, 'left' (5') flank, 'right' (3') flank and sequence of the PCR product (which together span up to around 150-300 bp). We did not receive the associated scaffolds and contigs from where this data was obtained. We have empirically validated dozens of resulting loci so far.
The problem is I want to retrieve an assembly wich contains all the loci from the table. BGI sent us some of the alleged parameters (such as kmer=47), but we can't generate an assembly (using SOAPdenovo 2) that yield all (or even most) of the loci as described in BGI's table. I have not succeed to get more help from them on that issue.
Parameters used:
SOAPdenovo all -K 47 -F -R -D -d -M 1 -s lib.cfg
lib.cfg
avg_ins=500
reverse_seq=0
asm_flags=3
rank=1
pair_num_cutoff=3
map_len=32
So my question is: is it possible to independently obtain the very same assembly they originally generated?
What can I do to get closer to this in case we are not 100% sure on the original parameters.
Thanks in advance and best wishes