[mira_talk] Re: Where is my assembly at?
- From: Visam Gültekin <teutara@xxxxxxxxxxx>
- To: Mira Talk <mira_talk@xxxxxxxxxxxxx>
- Date: Sun, 25 Sep 2011 16:55:48 +0300
Thank you for rapid reply. With comparison to newbler and celera, mira is the
absolute winner ( for data i have).
It is completely off the topic, but what if i have some EST sequences that i do
not know which technology has been used for sequencing(Sanger-454..),
i did some test with mira --job=est,.. for Sanger and 454 independently. There
seems to be no difference b/w results. Any suggestions?
From: bach@xxxxxxxxxxxx
To: mira_talk@xxxxxxxxxxxxx
Subject: [mira_talk] Re: Where is my assembly at?
Date: Sun, 25 Sep 2011 15:34:33 +0200
On Sunday 25 September 2011 15:23:23 Visam Gültekin wrote:
> It might be inappropriate question, but why do you think MIRA is not
> suitable for plant genomes?
>
> I have a 454 data for two plants ( 454 titanium, not paired end) and MIRA
> is doing good, I am awaiting reply with great curiosity
Oh ... my bad. Wrong wording. In fact, I also to plant RNASeq de-novo assembly
now (oh the fun).
Let me rephrase my statement.
I think that MIRA is suited for de-novo assemblies with
- Sanger reads: 5 to 10 million
- 454 reads: 5 to 20 million
- Ion Torrent reads: 5 to 20 million
- Solexa reads: 15 to 30 million (perhaps 50), provided the coverage is <= 80x
to 100x
PacBio is not yet in the list above, I need more data to give good estimates.
The above is usually OK for prokaryots and most lower eukaryots, however most
plants and higher eukaryotes need larger data sets for de-novo.
Artemus had 165 million Solexas for his assembly ... and there I have the
distinct feeling that MIRA will choke at one point or another.
Best,
Bastien
Other related posts:
