Question: Augustus and SNAP training in Maker-p pipeline
gravatar for ashaneev07
15 months ago by
ashaneev0720 wrote:

I'm planning to do a de novo plant genome annotation of using Maker-p software. I've completed the installation steps as per the reference ( Also did a trial run and everything seems to be good. My question is that,

1) Is it mandatory to train the augustus or snap for de novo gene prediction?

2) I've seen some articles related with this incorporated repeatrunner in addition to repeatmasker. but, i didn't install it with Maker-p. Is it affect the maker annotation quality?

Please give me a clarification regarding this....Thank you

ADD COMMENTlink modified 15 months ago by lieven.sterck9.5k • written 15 months ago by ashaneev0720
gravatar for lieven.sterck
15 months ago by
VIB, Ghent, Belgium
lieven.sterck9.5k wrote:

1) it is not absolutely necessary (from a technical point of view to train augustus/snap for de novo prediction but the quality of the resulting gene annotations should drastically increase when you do train it. If a trained version of augustus for your species should exist already you can of course use that one.

2) more or less same answer as above. It's not absolutely necessary but oyu might increase the quality of the results when you do it anyway. If you simply run RepeatMasker you will mask most of the transposons/repeats but quite often there are species specific variants of the TEs which might not get masked by using the default libraries of RepeatMasker. Building a library for your species specifically (== repeatrunner or repeatmodeler tools) will identify and mask much more potential repeats and will thus result in a much cleaner annotation.

ADD COMMENTlink written 15 months ago by lieven.sterck9.5k

Thank you so much for your answer. is it necessary to train both the gene predictors or any of one?

ADD REPLYlink modified 15 months ago • written 15 months ago by ashaneev0720

Here a tutorial to train Augustus.

ADD REPLYlink written 15 months ago by Juke345.0k

Hii.. I have few more doubts regarding the annotation with maker. I found that some articles describes Maker annotation has two repeat masking steps- through repeat masker and repeat runner. I've installed repeat masker program as mentioned in ( When i'm trying to install repeatrunner, they are asking CGL support. But i couldn't install it in my system. can i install repeatrunner with CGL? Does it work? otherwise repeatrunner is essential for maker-p? how can i incorporate it into maker exe file? Could u please have a look into this problem and help me out of this. Thank you so much in advance.

ADD REPLYlink written 14 months ago by ashaneev0720

What do you mean installing repeatrunner? I thought it was part of MAKER itself. Could you give us a link working to the installation procedure you are following, or copy past it here?

ADD REPLYlink written 14 months ago by Juke345.0k

Yes.offcource. The following tutorial was used for the installation of maker. But that doesn't include repeatrunner. So i downloaded it from however I couldn't install it because of the supporting software cgl. From some reference articles discussed about the repeat runner sw in addition with repeatmasker for masking repeats. This is my very first annotation work with maker p. So that I'm little bit confused too. I need a good quality annotation result. I don't know that u get my thoughts as well.! Need ur advice.. Please consider...

ADD REPLYlink written 14 months ago by ashaneev0720

again same as above: no, but the more/better you train them the better the end result should get.

If you train one, you show awareness that you should train them so you're doing a good job, if you train all of them you'll do an excellent job (with the downside being time and effort you put into it of course)

ADD REPLYlink written 15 months ago by lieven.sterck9.5k
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1114 users visited in the last hour