Metagenome assembly

[Yue Xu]

For the assembly of paired-end only Illumina data, I like to use ABySS assembler. But if the metagenome is too complicated, I agree with the previous post that both 25 M and 69 M reads are just to scratch the surface. Using different assemblers won't make signficant difference in terms of the number of contigs or n50.

If the purpose is just to recover genes from the metagenome, paired-end only Illumina data is useful to uncover genes except for those that suffer from strain variations. But to increase the integraty of the assembly dramatically (increase n50), mate-pair data with long inserts can significantly increase scaffolding performance. With some programs to resolve some gaps within scaffolds, the assembly can be improved further.

Hi, what are there tools for closing gaps in metagenomes? such as, Bambus 2, and what?Thank you very much.

[rhinoceros]

Metamos (with SOAPdenovo) appears to be a rather decent assembler although I have no personal experience with it..

http://genomebiology.com/content/pdf/gb-2013-14-1-r2.pdf

[Shuiquan]

For the assembly of paired-end only Illumina data, I like to use ABySS assembler. But if the metagenome is too complicated, I agree with the previous post that both 25 M and 69 M reads are just to scratch the surface. Using different assemblers won't make signficant difference in terms of the number of contigs or n50.

If the purpose is just to recover genes from the metagenome, paired-end only Illumina data is useful to uncover genes except for those that suffer from strain variations. But to increase the integraty of the assembly dramatically (increase n50), mate-pair data with long inserts can significantly increase scaffolding performance. With some programs to resolve some gaps within scaffolds, the assembly can be improved further.

[BIOin]

Currently we are finalizing our Assembly. It will be a great help you share your Ideas..

[yzzhang]

I tried to assembly a metagenome (plant endophyte, the plant genome is not avaiable now) uing ILLUMINA hiseq 2000 2*100 reads too, my data has 69 M paired end reads, 9.9 Billion bases. I assemblied these reads using CLC genomic workbench, and got 770 thousands contigs. I am working on these contigs now. How do you deal with your so many contigs? Could we share our idears>

thanks for the reply.
yes my data is complex(animal rumen), my data set Illumina 25M HiSeq 2000 2x100,

I just started using meta-velvet to assemble high quality metagenome data. I tried running meta-velvet with a k-mer of 45, after the assembly is finished and I look at the output file "meta-velvetg.contigs.fa" got 1128469 contigs with max contig length 31758 bp and N50 190.
Should i have to consider this assembly or need to run more Kmers...
Please give me suggestions on assemblers to be use

thanks for the reply.
yes my data is complex(animal rumen), my data set Illumina 25M HiSeq 2000 2x100,

I just started using meta-velvet to assemble high quality metagenome data. I tried running meta-velvet with a k-mer of 45, after the assembly is finished and I look at the output file "meta-velvetg.contigs.fa" got 1128469 contigs with max contig length 31758 bp and N50 190.
Should i have to consider this assembly or need to run more Kmers...
Please give me suggestions on assemblers to be use

[BIOin]

thanks for the reply.
yes my data is complex(animal rumen), my data set Illumina 25M HiSeq 2000 2x100,

I just started using meta-velvet to assemble high quality metagenome data. I tried running meta-velvet with a k-mer of 45, after the assembly is finished and I look at the output file "meta-velvetg.contigs.fa" got 1128469 contigs with max contig length 31758 bp and N50 190.
Should i have to consider this assembly or need to run more Kmers...
Please give me suggestions on assemblers to be use

[krobison]

i want to assemble 25 million reads. i am getting varying results with different assemblers.

With any dataset you will get different results with different assemblers, and even different results with different parameter settings of the same assembler, and different results with the same assembler and same parameters but different pre-processing steps.

For a metagenome, the complexity can vary depending on your sample. If you had a very complex sample, 25M reads (platform? paired end? read length?) is probably barely scratching the surface -- 25M 2x100 Illumina reads is only 5Gb, which isn't gigantic if you have a diverse sample.

[BIOin]

i want to assemble 25 million reads. i am getting varying results with different assemblers.

[severin]

Assembly

Many contigs can be assembled in a metagenome.

[RickBioinf]

there are too many variables to answer your question can you be more specific?

[BIOin]

How many contigs one can get after metagenome assembly?