how to do illumina sequencing assembly

[paa6]

I am very new to assembly. I am using geneious and velvet softwares. I have one reads file which is in fasta format and it is from illumina sequencing, size is 608 mb.

now, anybody can suggest me about the workflow..//

[mastal]

read the velvet manual

http://www.ebi.ac.uk/~zerbino/velvet/


it has many examples of commands you can run.

[GenoMax]

@paa6 are you certain the file is indeed from illumina sequencing this time

Generally one would not get a fasta format file (fastq is the norm) but your sequence provider may have already converted the original data file to fasta. Ideally you should start with the original fastq data for any analysis. The quality values are used by certain programs during analysis.

[paa6]

@paa6 are you certain the file is indeed from illumina sequencing this time

Generally one would not get a fasta format file (fastq is the norm) but your sequence provider may have already converted the original data file to fasta. Ideally you should start with the original fastq data for any analysis. The quality values are used by certain programs during analysis.

I think so u r right, n today I have discussed this thing with my PI but he didn't understand. but luckile I have managed to find one file...which is in .txt but after analysis I have seen that it's fastq file. Because I did analysis with illumina sequence identifier.

So, now I should use fastq file and do quality clipping (means trimming ends)???

[GenoMax]

Did you QC with FastQC first to see what your dataset looks like? Then think about trimming if needed.

[paa6]

Did you QC with FastQC first to see what your dataset looks like? Then think about trimming if needed.

OHH ok I will do that first good...thanks a lot...

[mchizar]

If you are doing de novo assemblies, it is essential to have the fastq files so that the low quality bases can be trimmed prior to assembly. If you are a newbie to de novo assembly, you will also want to use a software program that is easy to use, runs on Windows and Mac and produces an output assembly file that you can easily visualize and make edits, etc. DNASTAR's Genomic suite provides this and you can try out a free demo for 30 days.

[paa6]

If you are doing de novo assemblies, it is essential to have the fastq files so that the low quality bases can be trimmed prior to assembly. If you are a newbie to de novo assembly, you will also want to use a software program that is easy to use, runs on Windows and Mac and produces an output assembly file that you can easily visualize and make edits, etc. DNASTAR's Genomic suite provides this and you can try out a free demo for 30 days.

Finally my supervisor provided file in fasta format (illumina sequencing). I run the assembly on that file using geneious software. And luckily I have got 290 contigs and consensus sequence too. Now, what should I do further??

[rhinoceros]

Check out the assembly pipeline they use at JGI. They do multiple assemblies with SoapDenovo and then in the end merge all the assemblies for a final assembly. It's IMO by far the best way to do assembly as kmer settings affect tremendously what gets assembled on given run..

[mchizar]

that is the challenge with de novo assemblies. There are a lot of good de novo assemblers; Velvet, SOAP, SeqMan NGen, etc., but there are very few programs that provide a good interface for working with the contigs that are generated. If your intention is to order the contigs into scaffolds and close some (or all) of the gaps and then annotate, you will need good genome finishing software. The best program that I know of for this is DNASTAR's SeqMan Pro - they have been doing de novo genome software for 20+ years. There is a 45 minute webinar on their website that shows the genome finishing tools in action. I am not aware of any open source programs that can do this without an enormous amount of manual effort? Maybe someone else can comment?

[paa6]

Check out the assembly pipeline they use at JGI. They do multiple assemblies with SoapDenovo and then in the end merge all the assemblies for a final assembly. It's IMO by far the best way to do assembly as kmer settings affect tremendously what gets assembled on given run..

Thanks I will try this...

[paa6]

that is the challenge with de novo assemblies. There are a lot of good de novo assemblers; Velvet, SOAP, SeqMan NGen, etc., but there are very few programs that provide a good interface for working with the contigs that are generated. If your intention is to order the contigs into scaffolds and close some (or all) of the gaps and then annotate, you will need good genome finishing software. The best program that I know of for this is DNASTAR's SeqMan Pro - they have been doing de novo genome software for 20+ years. There is a 45 minute webinar on their website that shows the genome finishing tools in action. I am not aware of any open source programs that can do this without an enormous amount of manual effort? Maybe someone else can comment?

I will discuss this with my supervisor...