ssujun02.arb: ARB-database of processed small subunit rRNA sequences retrieved from the EBI, RDP, and Antwerp databases. Only seqeunces comprising at least 1000 nucleotides (excluding up- and downstream regions) are included. Usually, the primary structure data are taken from EBI, whereas the additional information is merged from all sources (tagged by [EBI], [RDP], [DEW]). The inclusion of eucarya sequences containing insertions/introns resulted in an alignment lenght of several thousand positons. To hide the many gaps use the respective tools of the editor (editor manual at arb-home.de) The tree_tre_1400_pub_may02 is an ARB-parsimony generated tree based upon (alomst) complete sequences only.It is evaluated ('global optimization') at the phyla level using phylum specific 50% filters ('filter by base frequency'). These filters are included as SAIs (alphaproteobacteria_rr5_may02: Alphaproteobacteria, columns containig more gaps (.-) than nucleotides removed, 50% conservation, May 02) The tree_tre_1000_pub_may02 is based upon tree_tre_1400_pub_may02. The partial sequences were added using the termini (removes up- und downstream regions) and domain specific gap filters (gap_99_eucarya_may02: removes columns at which nucleotides are present in only 1% of all full sequences from eucarya) and applying the 'add marked species' function of ARB_parsimony without further tree optimization. The alignment of hihghly variable regions or insertions of eucarya sequences still needs to be improved in many cases. (Remove those regions by appropriate filters for treeing!). How to merge your data and the new release: The file 6spring2001_2_ssujun02 contains the information on position and number of additional columns in ssujun02.arb in comparison with 6spring2001.arb. Use the ARB_merge tool (to be started from the introductory window) to find 'species' which are in your database (import as databse I), however, not in ssujun02.arb (import as database II). First synchronize the 'names' of both databses ('check names') then search for 'species' only present in database I (define 'name' as the database field to be searched and empty the search string window), mark the listed 'species', save database I. Open database I and export ('file > export') the marked 'species' to a new (smaller) database. Insert additional columns into this database ('Sequence > Insert_Delete column') according to the information in 6spring2001_2_ssujun02 (if your data were inserted in 6spring2001.arb. otherwise you have to find the positions by comparing reference sequences of your database and ssujun02.arb) by starting with the highest positon number. Alternatively you may use the 'preserve alignment' function of the merge tool: check the box and specify a number of 'names' for reference sequences in both databases. This works pretty good if you transfer groups of related 'species' and specify related references. The third alternative is to transfer (merge tool) all 'species' missing in database II. Save and open database II, edit the transfered sequences and some reference sequences (originally contained in database II) as templates, select the transferred sequences (see editor manual at arb-home.de), create a new group and insert the needed gaps in the consensus sequence of the new group. !!Take into consideration that inserting '-' in the consensus results in an additional '-' in all sequences of the group if a base symbol or '-' are the nearest neighbors. If '.' is one of the nearest neighbors this will be changed to '-' and no additional symbol will be inserted. This may disturb your alignment. (Will be changed in the next ARB-software release) Currently, the only way to circumvent this is to replace all '.' by '-' in that group (see editor manual).