leBIBI PPF                      Prokaryotes Phylogeny Facility - PPF v 1.1               News

SSU/LSU leBIBI_DB: mk41.1 (minor improvements of mk41) is available (mk41.1_procaryota...)
RefSeq sequences are tagged R. A TR tag corresponds to a Type strain (T) that is also a RefSeq sequence. For species where T (type) or R are missing, the reference is noted U (only one sequence found) or C (chosen best representative)

Query construction
QUERY with possible recruitment of the most similar sequences One or more accession nos. (ex: #AB681800 @superstringent %tag) OR name(s) (ex #Sodalis_praecaptivus @superstringent %notag)
OR sequences in fasta format (no extraction to do => no database to indicate, no tag to choose)
DB selection for extraction: @database with @[lax, stringent, TS-stringent, superstringent, genuslevel]. Tagging: %[tag, notag]
When submitting a mix of the query styles, the # containing selection rule must be AT THE BEGINNING
Large phylogenies (N>1000) may be long or very long to compute. PPF is more adapted to study relatively close phylogenetic relationships.
Blast parameters used for recruitment of similar sequences
Max # of BLAST hits to align :  (set #0 to suppress recruitment)
E-value [upper bound on the expected frequency of chance occurrence of a set of HSPs (-e)]
Sequence Database used for the recruitment of similar sequences (last updated 2020/Oct/02) Note that the "lax" database is NOT recommended to recruit proximal sequences to the query, please use it only with a limited number of sequences !
Supplemental Query without recruitment With the same grammar as above you may submit a request but the most similar sequences will not be recruited.
This may be usefull to recognize some sequences as an external group.
Never use here the "lax" database as target for extraction ("@lax"), this huge database is not strand-homogeneous and must be treated on the fly to find the strand.


User-given Id. : Input sequences
Max length of sequence names: [The shorter the name the cleaner the tree, pay attention to the remaining discrimination power]
Post-BLAST processing
Build phylogenetic tree
Build tree for each sequence [You may use "Yes" if your input data contain phylogenetically distant sequences]

Analysis Parameters
Program Accuracy     Depending the program, the "accurency oriented" option may give the optimal alignment (and is always very slow).
Selection of phylogenetically informative regions
Algorithm Use BMGE
Use Gblocks
No post-processing of the alignment (better if closely related sequences)
Lower stringency
Low stringency
Median stringency
High stringency
Higher stringency

Results will not be conserved for more than 18 hours, if needed please upload your working directory by using the "All files" link and selecting the leBIBIdatas.tar file