CLC genomics server for assembling large genomes

[ngs_agd]

Has anybody tried using CLCBio Genomics server for assembling large genomes (Panda, Human etc)?

We have interest in assembling large genomes with small amount of RAM. Tried using CLCBio server based on their claims, but we were not able to do de novo assembly of panda (from Li et al Nat Gen 2010) and Human (from SRA) reads generated on Illumina platform.
Our server configuration is 128 GB RAM and 2 Intel quad cores.
We ran the assembly (both contig generation and mapping) with all default parameters.

We would like to hear if anybody has had similar or maybe better experience.

[maubp]

I know some of my colleagues have had some success with the "CLCBio Assembly Cell" (or whatever their command line assembly tool is called) with plant genomes. Success meaning getting any useful contigs from a de novo assemblies, compared to most tools which needed too much RAM to do anything.

Sorry I can't be more detailed.

[NextGenSeq]

I was able to de novo assemble a 2GB genome pretty well with 24GB of RAM.

[ngs_agd]

I was able to de novo assemble a 2GB genome pretty well with 24GB of RAM.

Can you please elaborate on this? What was the organism, how much data did you provide, the platform on which the data was generated, how long it took to complete the assembly etc. Greatly appreicate your response.

[NextGenSeq]

It took a week with 350 million reads. Anymore than this you need more RAM. It's not published yet so can't say the organism besides it's vertebrate.