xTea on plant species

[agolicz]

Hi,
We wanted to try xTea on a plant species. Is that possible or does it require a human reference?

Agnieszka

[simoncchu]

Hi, the current repeat library is only prepared for human. You need to prepare a library for the plant species that you want to work on. Generally, you need to know what type (family) of repeats you are working on, the consensus sequence of them, the reference genome of the species, and also the repeat annotation (from RepeatMasker or other tools). With this, we can generate the library for the plant species and run on the alignments. I didn't try this before. I think it should work, but need extra effort for library prepare.

[agolicz]

Thanks! We will give it a try see how it goes.

[mnshgl0110]

It would be really helpful if there is a guide on how to generate repeat library for non-humans.

@agolicz did you manage to run xTea on plants? Could you please share how did it go?

[agolicz]

Unfortunately I have not had the time yet. Hopefully at the beginning of next year.
EDTA https://github.com/oushujun/EDTA, seems like a good candidate for plant repeat library creation.

[simoncchu]

I put a readme to prepare the repeat library here: https://github.com/parklab/xTea/tree/master/xtea/rep_lib_prep. Would you like to have a try? Note, xTea only works for TE insertions of known type, and need the repatmasker annotation of the TEs you are interested, and also a consensus sequence.

[agolicz]

Would it also be possible to add fasta header format (>xxx) for the TE library and the repeatmasker command to be used with that library?

[simoncchu]

What do you mean by fasta header format? For which file?
The repeatmasker output format is explained here: https://www.repeatmasker.org/webrepeatmaskerhelp.html. To run RepeatMasker on your assembled genome, you need to construct a consensus sequence library, and then feed in RepeatMasker (will call blast) to do the annotation. You can run tools like repeatscout to construct the consensus from the assembled genome. You can check the parameters by running RepeatMasker --help. If the species you are working on has some published reference genome, with big chance other people had annotated the genome, and you can directly use those.

[agolicz]

For the file:
TE-type.consensus.fa and TE_copies_with_flank.fa
Per repeat masker documentation: https://www.animalgenome.org/bioinfo/resources/manuals/RepeatMasker.html
The recommended format for IDs in a custom library is:

repeatname#class/subclass
or simply
repeatname#class

I just wanted to confirm that xTea expects the same.

For running pipelines like that on non-model species it is very helpful to have toy datasets, so we can ensure everything is formatted as expected. Some formatting conventions are not the same for human/animal/plant genomics.

[simoncchu]

You only need to extract the TE type you wanted to work on (each type separately). For example, if you have a repeatmasker output for the whole genome named species_rmsk.out, and you are interested in say LINE1, then you could run grep "LINE1" species_rmsk.out > species_rmsk_L1.out.

When generate the TE_copies_with_flank.fa, you only need to feed in the full length copies, so you need to select based on length from the generated pecies_rmsk_L1.out.

I am thinking of having a script to automatically generate this, but it's not easy to have a fix mode. Different species/TEs are of different length and different ids (some are customized set).

[agolicz]

Ok, thanks that makes sense. I will try to give it a try in January.

[adriaaanarcillo]

Hello, @agolicz! I would like to ask if you have successfully used x-Tea on plants already? If so, how did it go? Thanks.

[DR-genomics]

Hello,
I tried to create a plant repeat library using your instructions given in: https://github.com/parklab/xTea/tree/master/xtea/rep_lib_prep. However, I received an error: xtea: error: no such option: -P

And I don't see -P option in the xtea help page as well. What does -P stand for here?

Thanks!

[simoncchu]

Could you try again by replacing xtea with python full-path/x_TEA_main.py ?

[DR-genomics]

I tried the following and got this error:
python xtea/x_TEA_main.py -P -K -p ./ -r ../refgenome.fasta -a RMasker.out -o /home/xTea/TE_copies_with_flank.fa -e 100

Traceback (most recent call last):
File "xtea/x_TEA_main.py", line 345, in
x_annotation.load_rmsk_annotation()
File "/gpfs20/scratch/dramacha/xTea/xtea/x_annotation.py", line 241, in load_rmsk_annotation
start_pos = int(fields[5])
ValueError: invalid literal for int() with base 10: 'position'