Error correction in sequence Assembly

Question

[midox]

hello,
Is that the error correction algorithms like Coral and Echo use a reference genome for the correction of errors? or they use the Denovo method?

Thank you for your reply

[SES]

Are you asking about correcting a genome assembly or correcting unassembled sequence reads? It might help to expand your post to indicate these things and whether or not you are interested only in the two programs you mentioned.

[midox]

I found these two algorithms of error correction.

I have unassembled reads and I search in errors correction algorithms to correct these reads (based on alignment or MSA).

[lh3]

See here. There are also two review papers on that page.

Answer 1

[rtliu]

A recent review paper: Correcting Illumina data

Key Points

• Error correction is essential for improving Illumina sequencing data before using it in applications.
• Existing correcting programs are effective, recovering an important amount of erroneous data, yet significant room for improvement remains.
• Recommended programs:
  • HiSeq data: BLESS, Musket, RACER and SGA
    • MiSeq data: RACER
    • Human data: Musket, RACER and SGA.

[lh3]

For human data, the best tools are lighter and the latest bless. The old bless evaluated in the paper wasn't very good.

Answer 2

[Neilfws]

Why not read the papers and figure that out yourself? Here they are.

Coral - Correcting errors in short reads by multiple alignments

ECHO: A reference-free short-read error correction algorithm

Let us know what you find.

[midox]

Thank you for articles.

In ECHO and CORAL without reference sequence they use the DeNovo method.

Do you know other error correction algorithms? I searched and did not find.

[Brian Bushnell]

BBTools has an error correction tool, BBNorm (which does denovo normalization and/or error-correction). It's run like this: ecc.sh in=reads.fastq out=corrected.fastq.