MQ team programs

Modelling DynW Rfit

Identifying BIBI webtool (Bacteria and Archaea Identification) BIBI Help and Description

wikis admin (restricted) DynW MQwiki

Bibi Le Submission

The main leBIBI server is usually working

here : [1]

If leBIBI main server is down

Except if this a a network problem here, you may try the backup server : [2]

BIBIle or leBIBI

The name has no importance, BIBI is but this "new BIBI" is based on a few Python CGI programs and fully html, You may use it inside a browser without java script for viewing trees and alignments. The appearence is that of an enligthened program, even if the code is complex. The use is also simplified. The name of this BIBI evolution is officially BIBIle for "ligth edition", this is also a remembering of well known commercial programs bearing this "le" suffix.

For the impatients

Simply press the "run now" button, without any change and then go to the Bibi Le Results help pages.
Note that the presentation and the options are now different (july 2010).

Learning how to work with leBIBI

When you enter Bibi Le, you get this view with a query sequence area, a DataBases? selection and running conditions area, a short description pannel.

Getting test sequences

If you dont have any test sequence, here is the way to get one :

You can select a gene, and then ask for test sequences. A new (plain) page opens and you can select and copy a given sequence.
Note that we need a nucleotidic sequence in Fasta Format !

>This is an exemple
GGTGTGTCCATCGCCAAGGAGATCGAGCTGGAGGACCCCTAC
GAGAAGATCGGCGCCGAGCTGGTCAAAGAGGTCGCCAAGAAG
ACCGACGATGTCGCCGGTGACGGCACCACCACCGCTACCGTG
CTCGCCCAGGCGCTGGTGCGCGAGGGCCTGCGCAACGTCGCG
GCCGGTGCGAACCCGCTGGGCCTGAAGCGCGGCATCGAGAAG
GCCGTCGAGGCCGTCACGGCCAAGCTGCTCGACACCGCCAAG
GAGGTCGAGACCAAGGAGCAGATCGCTGCCACCGCGGCCATC
TCCGCCGGCGACGCGTCCATCGGTGAGCTGATCGCCGAGGCC
ATGGACAAGGTCGGCAAGGAAGGCGTCATCACCGTCGAGGAG
AGCAACACCTTCGGCCTCCAGCTGGAGCTGACCGAGGGTATG

The length of the lines does'nt matter and you may even use the nucleotidic sequence issued from a BanKFormat like here:

>this is a strange way to do
421 ggacgaagct tttgtgacgg tacctgtata agaagcaccg gctaactacg tgccagcagc
481 cgcggtaata cgtagggtgc gagcgttgtc cggaattact gggcgtaaag agctcgtagg
541 tggtttgtcg cgtcgtctgt gaaattccgg ggcttaactc cgggcgtgca ggcgatacgg
601 gcataacttg agtgctgtag gggagactgg aattcctggt gtagcggtgg aatgcgcaga
661 tatcaggagg aacaccgatg gcgaaggcag gtctctgggc agtaactgac gctgaggagc
721 gaaagcatgg ggagcgaaca ggattagata ccctggtagt ccatgccgta aacggtgggc
781 gctaggtgtg agggtcttcc acgactttcg tgccgtagct aacgcattaa gcgccccgcc

Of course Bibi Le will warn you about an unexpected number of unknown positions, but the result will be correct !

Copying the query in the query box

Copy your query sequence (it may be one of the test sequences) in the form here :

You have now to fix the value of some variables. Only the following is mandatory : Bibi Le cannot guess the gene you are using.

Selecting the relevant sequence data base

This is the only mandatory information, if you dont select your Data Base, leBibi will uses the default Bacteria_SSU_RDNA (16S) Data Base. You may get back irrelevant results or an information on the error.

Names of sequences databases are indicating the scope (Bacteria, Actinobacteria ...), the gene (ex: SSU-rDNA-16S) and the taxonomic and bacteria nomenclature compliance (stringent) or non-compliance (lax)
the stringent choice led to a taxonomically relevant identification. Type strains are identified (but some may be missing due to incomplete descriptions in Gen Bank)

Note : now the databases are actualized every 3 months

Selecting the number of sequences to compare

Now select the number of sequence to compare, the default is 30, but it may be better to choose 50 or even 75. The highest the number, the slowest the result. Use of more than 75 is for special situations (plenty of identical strains giving a flat tree for instance, in this case, please consider using the TS data base if you are using SSU_rDNA).

Selecting inclusion or exclusion of gaps

The best tree is obtained using all the information, thus the default is to include gaps. In special cases you may choose to toss gaps, only the positions on the sequence without any gap will be taken into account, this led to less informative trees.

Selecting the alignment speed

Muscle is used for alignment, usually its accuracy is good, even with the fastest option. Of course the alignment may be improven by using more accurate choices (but you will also select a slower process).

Selecting the cleaning level

The default is "medium". The alignment cleaner removes low quality ends to some extention. These tails are causes of erroneous position in the tree. The lower level does cleans only the N-N tails and this has no effect on the tree. The upper levels enables the suppression of the N-rich parts on both side of each sequence of the alignment. Trees may be substantifically improved in some cases by choosing the higher level.

Now run

Click !

And then go to Bibi Le Results help page if necessary

Site powered by MonkeyWiki