How to use the pipeline?

Official URLs

Home:

• http://www.clermont.inra.fr/triannot

Direct access to the submission window with login/password :

• http://urgi.versailles.inra.fr/triannot/

Input BAC sequence

You can upload, each time, a multi-fasta file up to 10 sequences maximum. The minimum and maximum size of each sequence is respectively >10 kbpand <3 Mb. Case insensitive. There is no restriction concerning the number of submission since the pipeline uses a queuing list for sequence submission.

Example :
>Contig385B22 from BAC T. aestivum BAC library Pool A
 TTTCTCTTTGGGATAATTAGATTTATGCCCCTAGTTGTGTCCCACTCGTC
 TGTTTTACCCCTAATTCCCAAAAGTCACCAGTTCTGTCCAAATCACTTTC
 CTCCTCTTATGCTTTTGCCCTTTGACCGTTTGACCGTTAGTTTGAAAACT
 TCATAACTAATTCATACTAAATCAGAAAAATTCAAATAAGATACCAAAAT
 GTTCAGAAAAACATCACCTATATGCCAGTGTCATTTGCATCCATGAAAAA
 AGTGTTGGAAAGTGCCCATCTGAGTTTTAGCTCTCATGCTACCACCATGA

Avoid blank within the sequence.
The sequence characters should be A T G C, as well as IUB common characters for DNA sequence :
U (T), M (A C), R (A G), W (A T), S (C G), Y (C T), K (G T), B (C G T), D (A G T), H (A C T), V (A C G), X/N (A C G T).

Analysis time

In principle, the pipeline can be used to annotate full genomes. However for technical reasons and parallelization purposes, 10 sequences up to 3 Mb can be submitted online at once. As it would be cumbersome to annotate several Mb or Gb of sequence this way, the online access is more adapted to small scale analyses (i.e. BAC or small BAC contigs) in which the user can submit its sequence directly on the webpage (copy/paste or download) and start the analysis with a single click. The pipeline uses a queuing list for sequence submission. Therefore, the automatic structural and functional annotation process will depend of the queue length. In general, in this configuration, TriAnnot can deliver a BAC annotation in less than one hour depending of the cluster charge. For example: a default analysis of a 117 kb sequence containing 6 genes takes about 30 minutes.

A management screen is available (My Analysis) to check the status of your analysis (see figure below):

Submission of your own sequence

First of all, when you log in the first time you will have to fill the following screen: “My Profile”. This has to be done once.

On the TriAnnot pipeline analysis submission screen (see below) you first have to give a title to your analysis, and then choose a pipeline template. This pipeline template will define the receipt of your analysis by building the step.xml necessary to the pipeline. At present there are five pipeline templates:

• “Wheat default IWGSC Annotation” - a default analysis which has been optimized for the annotation of the wheat chromosome 3B (French ANR 3BSEQ project).
• “Rice default analysis” – a specific step.xml (template) has been written for the rice sequences – more suitable databanks combination and/or ab initio gene predictors. However, this template is not optimized as it has been done for wheat
• “Oak default analysis” – same as rice. However, for oak specific RNA-seq data are used to improve the gene prediction based using SIMsearch
• “Barley default analysis” – same as oak
• "Maize default analysis" - same as rice

Then, you paste or upload your sequence before you click on the “Submit analysis” button.

Remarks: Other templates could be proposed under request if necessary, especially if you want to use the TriAnnot pipeline for other species. In this case please, contact triannot-support@clermont.inra.fr.

Email & links

The pipeline automatically sends you an email when data are available for your sequence’s structural and functional annotation. Here an example:

The link will take you back to the TriAnnot management interface (My Analysis). Then, you have two possibilities:

1. have a quick look at the results using a graphical display such as GBrowse
2. download all your data to your own computer.

GBrowse graphical display

Using the “My Analysis” web page you can also display a graphical view of your annotation using the Genome Browser. Of course, the online GBrowse viewer cannot be considered and used as an editor for manual expertise.

By default and speed up the display only four tracks are shown:

• the sequence overview;
• the Gene overview;
• the “Structural & Functional Gene Annotation” (the TriAnnot predicted gene models) track
• the “05_RepeatMasker – TREPplus” track.

Of course you can add new tracks by using the “Select Tracks” tab. Be aware thanyou can’t keep your track configuration from one analysis to another. There is no way to save the current configuration!

Download all your data

Using the “My Analysis” web page you can download the analysis of your choice. You will recover like that all generated output files (gff, embl, align, etc … - see the “Output Files” paragraph below) to be used locally with your own graphical editors (Artemis/GenomeView/Apollo). Each embl, gff or align file is tagged with the TriAnnot step analysis number (seen Architecture of the pipeline). You can also recover the initial and masked sequence in FASTA format which has been submitted, and protein FASTA file (translated gene models).

With the TriAnnot management window you can always delete previous analysis.

Output Files

The TriAnnot analysis will generate several output files organized within 4 folders. (see Architecture of the pipeline for more details).

• BLAST results folder
  • EMBL folder
    • A number related to the step number: 06_BLASTN_*; 07_BLASTX_* and 12_BESTHIT_BLASTP_*
    • _* correspond to the databank used (see databanks)
    • Extension .embl
  • GFF folder
    • Same as above but with .gff extension

• EMBL folder
  • In each folder files are tagged and follows the following rules:
    • A number related to the step number
    • The type of programs i.e. REPATMASKER; AUGUSTUS; EXONERATE; EUGENE; GENEMODEL; SIMsearch; BESTHIT; TRNASCAN-SE; TRF; BLASTN; BLASTX
    • The databank used (see databanks). When no databank is used, just the type of program is displayed i.e. 16_TRF.embl. For ab initio gene prediction programs the matrix used is displayed i.e. 5_AUGUSTUS_wheat.embl
    • Extension .embl
  • Few examples for wheat:
    • Step1- tRNAscan
      • 1_TRNASCAN-SE.embl
    • Step2 -Transposable Elements annotation & masking / univec and E. coli contamination
      • 2_REPEATMASKER_Ecoli.embl
      • 2_REPEATMASKER_univec.embl
      • 2_REPEATMASKER_MIPSrepeatPoa.embl
    • Step3 -BLASTx against TREPprot
      • 3_BLASTX_TREPprot.embl
    • Step5 - ab initiogene prediction
      • 5_AUGUSTUS_wheat.embl
      • 5_FGENESH.embl
      • 5_GENEID.embl
    • Step6- BLASTn / Exonerate
      • 6_EXONERATE_cdsBDI.embl
      • 6_EXONERATE_cdsOSAirgsp.embl
      • 6_EXONERATE_rnaSeqWheat.embl
      • 6_EXONERATE_TAEugs.embl
      • etc ...
    • Step7- BLASTx/Exonerate
      • 7_EXONERATE_pepBDI.embl
      • 7_EXONERATE_pepZMA.embl
      • 7_EXONERATE_protTRIT.embl
      • 7_EXONERATE_SIMprot.embl
      • etc ...
    • Step8- Combiner
      • 8_EUGENE.embl
    • Step9- SIMsearch results
      • 8_SIMsearch_TRITfl_CAT01.embl
      • 8_SIMsearch_SIMnucWheat_CAT02.embl
      • 8_SIMsearch_MAGNmrna_CAT03.embl
    • Step10- Gene structure without functional annotation
      • 10_MERGE.embl
    • Step11 - Genestructure with functional annotation
      • 9_GENEMODEL.embl
        • This is the most important file since it gives the final gene structure and contains the functional annotation
    • Step12– BLASTp / Exonerate (Best Hit)
      • 12_EXONERATE_pepBDI.embl
      • 12_EXONERATE_protSAC.embl
    • Step13– InterProScan for protein domains identification and Gene Ontology tag
      • 13_INTERPROSCAN.embl
    • Step14 – BLASTn (CNSs)
      • 14_BLASTN_genoBDI.embl
      • 14_BLASTN_genoOSAirgsp.embl
      • 14_BLASTN_refSeqChloro.embl
      • 14_BLASTN_refSeqMito.embl
      • 14_BLASTN_tgacWGSAash.embl
      • etc ...
    • Step15-BLASTx (CNSs)
      • 15_BLASTX_SIMprot.embl
    • Step16–Microsatellite markers (SSRs)
      • 16_TRF.embl

• GFF folder
  • Exactly the same as above except that the file extension is .gff

• Other files folder (as an example)
  • Best hit alignment files with percentage coverage and identity, and missing or additional gaps over 9 amino acids
    • 12_BESTHIT_BLASTP_pepBDI.align
    • 12_BESTHIT_BLASTP_pepOSAirgsp.align
    • 12_BESTHIT_BLASTP_protTRIT.align
    • etc ...
  • Files results obtain with GTallymer
    • 4_GTtallymer_Cs1XOCC1.fplot
    • 4_GTtallymer_Cs1XOCC1.res  (to display the k-mer composition of repeat sequences)
  • There is also a tabulated file for TEs annotation
    • Global_XM_for_TE_RNA_Nmask.xm

• sequences folder
  • Initial fasta file submitted to TriAnnot
    • initial.seq
  • Several type of masked sequence (Ns / lower case)
    • RNA_Nmask.seq / RNA_LCmask.seq
    • TE_RNA_Nmask.seq / TE_RNA_LCmask.seq
    • Gene_TE_RNA_Nmask.seq
  • Protein sequences derived from TriAnnot gene model annotation. This file is important since it gives the final annotation based on protein sequences which can be used for further analysis. All proteins should start with a M (Methionine) and end with a star (*)
    • proteins.seq

Manual Annotation

After downloading the gff/embl files, and for manual expertise, you may use graphical genome annotation curation editor programs such as:

• ARTEMIS (Carver et al., 2008 Bioinformatics 24, 2672-2676)
  • In this case it is preferable to use the EMBL files

• GenomeView (Abeel et al., 2011 Nucleic Acids Research 2011;doi: 10.1093/nar/gkr995)
  • GenomeView needs a unique feature name to differentiate each track. Therefore each EMBL file is created with this constraint. Within the Databanks link we give for each EMBL file the feature name used with GenomeView (track list - see databanks)

• APOLLO (Lewis et al., 2002 Genome Biology 3)
  • In this case it is preferable to use the GFF files