8.4.066 - 16.09.2010 - Seqtools Configuration. - Minor correction of the automatic update function for seqtools.exe available on the Help/SEQtools Configuration.../ form. This update requires a new download and install of Seqtools to be effective.

Cheers

Soeren

8.4.050 - 20.08.2010 - QBlast function for batch blast search is repaired. - As a result of slight modifications of the QBlast server at NCBI the QBlast function in seqtools has not been working for a while. The code in this build of seqtools has been adjusted to handle these changes - and the function for batch blast search on the QBlast server now seems to work.

I have tested the function on some 600 nucleotide sequences of varying length and with the exception of a few of the sequences which consistently failed, the batch blast search on the QBlast server worked without problems.

Note that it is possible to set the waiting time for retrieval of results which causes the function to wait the specified number of minutes before re-submitting the sequence to the QBlast server. If a result is not received from the QBlast server after the second submission, the sequence is dropped and the function continues with the next sequence. Setting the wait period to 'Off' causes the function to drop the sequence without re-submitting it.

I would appreciate it very much if users who encounters problems with the QBlast batch search function contact me with a description of the problem.

Cheers

Soeren

8.4.045 - 27.03.2010 - A few - unimportant - adjustments of form layouts. - Somewhat to my surprise, the number of seqtools users is slowly but steadily increasing. Now being retired, I seldom use the program myself execpt for occasional testing of various functions included in seqtools. But I am pleased that other persons still find my program suite useful in their struggle with dna and protein sequences.

So far neither I nor other users have detected bugs for a very long time which may optimistically indicate that most bugs have been killed and the rest are located in places where nobody looks.

Cheers

Soeren

8.4.042 - 06.09.2008 - Bug in local blast search functions corrected. - Two quite serious bugs recently detected by Birgitt Oeser have been corrected in this version.

When doing local blastx searches - against single or multiple local databases - the option to search one or both strands of the query sequence was incorrectly disabled if the blastx  program was selected from the list of blast programs. The setting was correctly detected by the blastx program but the user was unable to alter it with a disabled checkbox.

The two bugs have been around for a very long time as this part of the seqtools code has not been altered for several years. I am really sorry if the bugs have generated problems for any users - although I know that my regrets don't help very much...

Thanks to Birgitt for drawing my attention to the problem.

Cheers

Soeren

8.4.040 - 17.06.2008 - New option added to header search function. - In response to a request by Patrick Gaffey an option has been added to the header search function. It is now possible to step through matching strings on the displayed header page by repeatedly clicking the "Find Next" command button on the small form handling header search results.

In addition two email options have been removed: The bugreport form and the possibility to send sequences by email directly from Seqtools.

Both functions relied on a third party dll (ANSMTP.DLL) which is erroneously identified as a Trojan by several anti-virus programs (see below). In the future you will have to use your normal email program for communicating this information.

Cheers

Soeren

8.4.032 - 15.01.2008 - New facility added to the Clustering function - In response to the request of Julien Sacré I have added options to remove from the current project either the sequences included in clusters or the unique sequences not included in clusters from the current project. The sequences are NOT deleted from the disk, just removed from the project.

The options are both activated after the selected clustering operation is completed. The new options are available for the three separate functions for clustering included in Seqtools (Blastn Score, Blastn Cluster and Blastn Alignments). The info line lists the number of sequences in clusters as well as the number unique sequences.

Comments are welcome...

I would also like to know if any users of Seqtools have installed the application on computers running under Windows Vista. I am still using Windows XP Pro and hesitate to setup a separate pc for testing the program under Vista. So please drop me a few words if you have any information worth sharing with me - and other users of Seqtools.

Cheers

Soeren

False positive virus / spyware warning - 11.11.2007 - There has been a few reports from users that a file included in the seqtools setup is a trojan: Trojan.WinSpy as shown in the AVG warning below.

The file referred to is a dll used by seqtools to handle emailing, "ansmtp.dll" and IS NOT A VIRUS. You can safely ignore the warning and include the file in the exceptions list of your AVG anti-spam/anti-virus program by clicking "Ignore and add to execptions" and checking "Use for all objects of this infection" as shown in the screenshot of the warning message below.



If you would like to read more about the ANSMPT component visit their website.

There is a newer version of the ansmtp component which does not have this problem. Unfortunately this requires buying a new - fairly expensive - license to the ansmtp component. The economy of the seqtools organisation (i.e., yours truly) does not allow this at the moment - as most users prefer to use the free student version.

There are a few bug fixes in the latest version (8.4.28) of seqtools: The auto-annotation function has been fixed and now works again.

Cheers

Soeren

What is your opinion about seqtools? - 13.05.2007 - As the author of seqtools I often speculate how users receive and use the program. I have invested a long time writing and supporting the program constantly trying to make the program as user friendly as possible.

The problem is that I usually only receive user reactions when something goes wrong - which unfortunately happens too often.

In attempt to get an impression of how seqtools works in the hands of its users in their daily work with protein and nucleotide sequences, I have made a questionnaire which allows you to convey your experience with and opinion of seqtools.

I hope you will take the time to write your personal mini-review of seqtools. Positive and critical comments are most welcome and will be carefully considered and used in future improvements of the program.

Fill out the seqtools questionnaire now and help me, yourself and other seqtools users to make seqtools a better program...

I would of course like to know who wrote the review, but in case you prefer not to reveal your nane it is quite all right. I will read your review with interest anyway.

Looking foreward to receiving your comments...

Cheers

Soeren

8.4.020 - 12.05.2007 - Virus problems with ANSMTP.dll - I have recently received two mails from users who reported that they got a warning from their antivirus progams (in one case: Avast Home Edition) telling them that the ansmtp.dll file is a virus.

Adminsystem Software is aware of the problem and has provided a workaround to fix the problem. Unfortunately this requires that I buy a new license to their product. This is quite expensive and in the light of the fact that most users of seqtools prefer to use the free students license not really a realistic option for me at this time. The ansmtp.dll handles all mailing operations in seqtools, i.e. really not a very important function of the program.

You can safely continue the installation of seqtools despite the warning that ansmtp.dll is a virus. It is NOT a virus, but a false positive of your antivirus program. You can read the two letters from ANSMTP describing the problem and the workaround here: letter 2005, letter 2007. I might add that I believe it would have been fair if the company provided a free solution to this problem. This certainly does not add to the reputation of "Adminsystem Software".

I have contacted Avast today about the ansmtp.dll file asking them to remove it from their list of virus signatures. They responded positively an hour later asking for a copy of the file. So it looks as if this problem - with Avast at least - is being resolved.

8.4.020 - 06.05.2007 - Major revision - I have received some user complaints after the release of the previous "upgrade" which deserve some comments and explanations. The main reason for a series of problems was a note from a user who mentioned that he no longer was able to load multi-sequence files generated with pervious versions of seqtools. At this time I had just routinely updated the NCBI programs - which I usually do, and have done for several years - when new releases become available.

The user send me copies of his old data ms-files of which one worked fine with the updated version of seqtools while the second ms-file included an empty sequence as well as several sequences with a large number of N's.

This has never created any problems before but with the "updated" new NCBI programs (version 2.2.16, 27-04-2007) the NCBI programs crashed (fortunately without killing seqtools at the same time) when they encountered an empty sequence and also erraticly died when processing a low quality sequence.

I spent a considersbly amount of time trying to fix this problem by modifying the seqtools code without much success and finally decided to revert to the 2005 version of the NCBI programs which I knew worked reliably. I might add that I have never detected any benefit from previous updates of the NCBI programs.

To avoid similar problems in the future, the NCBI programs now used by seqtools (release: June 5th. 2005) were renamed as follows:

blastall.exe -> st8_blastall.exe
formatdb.exe -> st8_formatdb.exe
fastcmd.exe ->st8_fastcmd.exe
bl2seq.exe -> st8_bl2seq.exe
blastclust.exe -> st8_blastclust.exe

and included in the auxiliary.exe file as well as in the installation file. As the filenames of the NCBI programs are hard coded in seqtools a re-installation is required to implement this change.

In the same update I included a few minor improvements including modifications of the clustering function which now stores the settings in the seqtools ini-file.

I am fully aware of the inconvenience this causes for users, but please remember that I am not introducing these changes to make life more difficult for you - even though this may seem so...

8.4.016 - 26.03.2007 - Major revision - This revision at the same time terminates the old seqtools 8.2 as well as the separate 8.3 version for teaching purposes. In the future only one version - 8.4 - of seqtools will be available.

For people using seqtools for classroom teaching the "school" version is now build into version 8.4. Typing a "s" while seqtools is loading, launches a limited version of the program in which only the very basal functions are available in menus and forms.

Users of the old version 8.2 can continue using 8.4 under the same conditions as before: No license fee but renewal of the registration every 60 days.

To help finance the development and maintenance of seqtools, senior scientists are encouraged to purchase a full license. Remember: Very few things in life are free - in most cases it just implies that somebody else is paying the build.

Regarding the program only a few changes have been made to this version. The search facilities on sequence lists have been rewritten and now works quite well. All forms have been edited to display correctly in Windows XP running the" XP style" - as opposed to  "Classical style".

A burning issue is becoming increasingly more relevant: Does seqtools run under Windows Vista? At the moment I don't know - and have no intentions of finding out myself. I am fed up with Windows updates which rarely adds functionallity but rather causes great disturbance until a number of service packs and updates have been released. At this point XP suit my needs and runs fairly stable.

A few users have reported installation problems concerning the MSINET.OCX file but I am uncertain at the moment whether this problem is solved or still persists.

If you have any information on this problem I would be very interested to know about it.

Send a note to S.W. Rasmussen:

8.3.052 - 13.10.2006 - Bug fix - Problems associated with closing the seqtools application have been fixed. This correction has also been implemented in seqtools version 8.2.108 for students.

I have noted that seqtools newsletters sometimes are rejected by spam filters. If you are interested in receiving information about seqtools, please mention this problem to the administrator of your mail account. It is quite important that it is possible for me to be able to reach you in cases where bugs in the program affect results generated by seqtools.

8.3.040 - 26.01.2006 - Bug fixes - (1) The active codon usage tabel can now be set and changed under the Preferences menu. (2) Calculating protein properties now also works when amino acid sequences contain lower-case letters. (3) A new routine has been implemented to handle e-mail based operations. This affects the bug-reporting facility under the Help menu and the sequence forwarding facility under the File menu.

Due to the new e-mail routine, the simple update function may not work. In this case please perform a complete de-novo installation of Seqtools.

8.3.037 - 11.12.2005 - A new "school edition" of Seqtools released - Several users have previously been employing Seqtools in elementary and high school teaching as well as in undergraduate university courses in bioinformatics. As the large number of functions and facilities in Seqtools often leads to confusion, a special version of Seqtools have been released.

In the "school edition" of Seqtools only very few, basic functions are available/visible to the students while the teacher (and perhaps capable/interested students) have access to the complete version of the program (simply by typing "full" while the program is loading). The complete version included with the "school edition" of Seqtools is identical to the "normal" edition 8.3 of Seqtools and will be upgraded in parallel with "normal" version.

Positive and/or negative comments from users may lead to changes in the functions included in the "limited" version of Seqtools. Depending on the interest in this option, a special edition of the user manual for "non-specialists" may be relevant.

8.3.033 - 23.10.2005 - The revision of the web manual is finally completed - During this revision all routines and functions in SEQtools were tested and corrected where necessary making this version of SEQtools probably the most stable and bug-free.

In this build a minor addition has been made to the "Search Project" Function making it possible to retrieve query strings from search data files (this option has previously been limited to the "Search Sequence" function). This enables you to search the entire project for sequences matching primer sequences created with the "Convert Project To Search Data File" under the "Tools" menu.

8.3.030 - 22.09.2005 - A serious bug in the batch blast function corrected - Thanks to Dag H. Coucheron a serious bug was detected which caused the parser for returned blast results from genbank/NCBI to ignore all description lines in the results file. This bug is corrected now and so is a number of less important bugs here and there. More bugs will probably be found as the manual revision/program testing progresses. So please be patient - and send me a note when/if you detect something weird.

The revision of the web manual is still in progress.

Sections:  1. Introduction, 2. Installation, 3. Features, 4.1. File, 4.2 Edit, 4.14 Preferences menus and 5. Primer functions have been revised. The rest will appear in the near future.

8.3.025 - 12.09.2005 - Trim raw sequence trace files rewritten... - This function suffered from several bugs - one preventing it from running under Windows XP operating systems. These bugs have been corrected (I hope...) and the functions now do what they were intended to do.

The revision of the web manual is slowly progressing. While revising the manual I realised that it was more out-of-date than I suspected. Sorry...

Sections:  1. Introduction, 2. Installation, 3. Features, 4.1. File menu and 5. Primer functions have been revised and section 4.2. Edit menu is on its way. The rest will appear in the near future.

8.3.021 - 30.05.2005 - Options to export sequence list as Excel spreadsheet... - This function has been a wish of many users for a long time. On my own pc it works perfectly well. However, I am not sure how the update and installation files handle the setup of the Excel part. Please let me know if you experience any problems updating seqtools to this build.

In addition to this function, a number of improvements is included in this update, the most important being a greatly increased stability of the batch blast search function. Recently almost 17,000 protein sequences were searched against genbank "nr" without any user intervention at all. The search took several days.

8.2.098 - 08.12.2004 - Several functions modified to increase speed... - The dot-matrix sequence comparison function has been improved to improve speed. The function is, howver, still not among the fastest, especially when sequence length exceeds some thousand bases. The sequence complementing function is also slightly improved.

This version of the executable is compiled as so-called p-code which according to several experts should generate faster code and - as an extra bonus - also results in a considerably smaller exe-file. If you experience any problems please let me know.

8.2.095-97 - 22.11.2004 - Clustering output forms redesigned... - The design of the forms used to display the results of sequence clustering has long been rather unsatisfactory. This has been changed in this update. The resulting clusters are now listed in a tree-view control which links directly to the form displaying members of the selected cluster.

Clicking the name of a cluster in the tree lists highlights all members of the cluster on the group form for easy selection of cluster members for alignment and/or merging.

Clicking a single member of a cluster in the tree also lists the entire cluster but only highlights the selected sequence on the group form.

As before, holding down the right mouse button on the group form displays the blast description lines for selected cluster member - provided of course that blast results are available.

The previous report form is now only used for printing and saving a summary of the clustering. The report form is opened from the main clustering form under the report menu. The print and save options are contained under the file menu of the report form.

Minor changes in this version includes improved setting of the number of descriptions, alignments and expect value cutoffs for blast searches on all blast search forms.

Finally, note that the collection of auxiliary components has been updated again. This time because of a recent upgrade of the chromatogram processing program from the Sanger Institute used by seqtools.

I have included a new option on the download page of the web which allows you to download the core seqtools installation and the auxiliary and emboss components as a single installation file. It is, however, still possible to update each component individually from the seqtools configuration form.

8.2.094 - 14.11.2004 - Bugs in clustering function corrected... - A bug preventing clustering large number of sequences has been corrected. Problems with loading and saving (with and without compression) large multi-sequence files have been corrected.

8.2.092 - 26.10.2004 - Bugs in save/export function corrected... - A bug preventing saving / exporting project data has been corrected.

8.2.091 - 04.10.2004 - Bugs in contig builder function, Windows bug!!... - A bug preventing selection of sequences to be merged is corrected. A new menu item has been added to the File menu enabeling you to close the current project without exiting seqtools.

IMPORTANT:  A serious bug in Windows 2000 and XP and possibly also in Win98 has recently (a few days ago) been detected. Unfortunately the bug also affects Visual Basic and programs written in this programming language. The bug affects the save-file dialog form (an integral Windows component used by Visual Basic), found in numerous places in seqtools. The bug occurrs when the user performs the following series of operations:

1. with the save-dialog form open: clicks an existing filename in the filelist, e.g. "MyFile.txt"
.


2. edits the selected filename in the text field to make it different from the original, e.g. from "MyFile.txt" to "NotMyFile.txt"
.


3. clicks on a "white / empty" region on the filelist displayed in the dialog form and clicks the "Save" button.

8.2.087 - 03.09.2004 - Bug in blast functions corrected... - Saving blast results either obtained by QBlast or by Blastcl3 searches failed when the "save results as separate files" option was selected. Also saving results as separate files in html format didn't work. Both problems are corrected in this revision.

8.2.085 - 20.08.2004 - Several minor bugs corrected... - Small changes here and there. Rename functions with more readable sequence name lists. Clicking the Exit key on the Registration form no longer causes a crash of seqtools.

8.2.083 - 23.07.2004 - Yet another attempt to improve auto-updating seqtools...  It has been unexpectedly difficult to get the program update facility to work properly due to differences in operating systems, server setup, firewalls etc. Several users have complained that seqtools freezes instead of downloading the update. This build addresses this problem.

8.2.082 - 20.07.2004 - New option for editing clusters... - It is now possible to display and edit the individual members of a cluster after clustering thanks to the suggestion by Florent Crepineau. Remember to let me know if you need functions or facilities not already included in seqtools. Often it is quite simple to modify seqtools to precisely match your needs.

8.2.080 - 08.07.2004 - Bugs in trimming raw sequences... - Several minor bugs in the function for trimming raw sequences under the Edit menu have been corrected in this build. User feed-back in general and related to this function in particular are highly welcome.

8.2.075 23.06.2004 - Bug in checking for an active IN connection... - It appears that some users still experience problems with the function testing if a live internet connection is available. This apparently trivial test has caused a lot of troubles in part due to bugs in the microsoft api function which behaves differently with different versions of windows (XP and win2000). When I test the function on my own PC's everything works as it should. Version 8.2.075 includes a second user option: "Ignore" which cancels internet connection checking by assuming that a live connection is available.

8.2.071 17.05.2004 - Bug in checking for an active IN connection... - Recent versions of seqtools contained a bug in the routine checking the internet connection status of the pc. The user was informed that no internet connection was present irrespectively of the connection state of the pc. This problem is corrected in version 8.2.071. Registered user have been informed in a recent newsletter. Other users should upgrade to version 8.2.071.

Two minor bugs affecting storing SAGE data and two-sequence comparison results in the sequence header have been corrected.

I will be out of office for a week by the end of May. Users with extended demo licenses will not be able to extend their licenses during this period.

8.2.051 24.03.2004 - More corrections... - Thanks to several user who have pointed out a few bugs and misbehaviours of seqtools this version, 8.2.051 may contain fewer bugs.

I have added visible icon-shortcuts to the most used options on the sequence list form - and have at the same time realised that icon design is not so easy. I may possibly be the only person finding a bit of logic in the relation between the icon and the underlying function...

Description line formatting - including the line substitution function - has been slightly modified/corrected and hopefully behave correctly now.

As you may have noticed the user comments form has changed a bit hopefully making it easier to understand and use.

Seqtools does not accept IUB symbols but converts them to N's. This function, however, with some load options, failed to perform this conversion but simply skipped IUB symbols - this has been corrected in version 8.2.051.

Large projects, heavily annotated tend to require a lot of disk space, especially inconvenient when users wish to mail projects to each other - or to me. Therefore I have included a compression facility (only working for seqtools formatted sequences) which reduces the size of a saved average project about five times. Compressed projects are recognised and automatically decompressed by seqtools when loading. The extension must be "fmz" for seqtools to decompress the ms file.

Finally, the context sensitive help has been updated - although there may still be (= are) unexplained items left. Remember, you are always welcome to contact me in case you have problems doing a particular job, have ideas for new functions or simply want to tell me that something works fine. Most people get inspired to write when something does not work - software developers on the other hand get happy when they receive good news...

10.03.2004 - Various corrections to several functions - Corrections have been made to many functions including: Blast blatch searching, description line formatting, management of information stored in sequence headers, facility for entering user comments, project save/export function. New functions include options to retrieve information from header description line sections generated with local database searches.

I wish to thank the many user who have taken the time to tell me about various problems.

30.01.2004 - QBlast batch blast function - I failed to realise that the maximum string length for an url in Microsoft IE is 2083 characters. As the first edition of the qblast function in seqtools used this method to submit blast requests to NCBI this resulted in submission failure if sequence length exceded some 1500 bases - which is obviously not satisfactory.

The modified qblast function in version 007 uses a different method of submission (http post) which do not have this limitation. Changing the submission method generated several other problems but at this point the function seems to work very well - also with long sequences. I wish again to thank the people at the blast helpdesk at NCBI for their friendly and competent way of handling users - also when they ask rather silly questions.

25.01.2004 - Menu icons added to selected forms - With the large number of functions and options in SEQtools it is sometimes difficult to locate a specific facility without having to read several menu item captions. Therefore I have added various icons to menu items on the main editor, the header form and the sequence listing form in version 8.2.

A few minor changes have been included in the version 8.2: The cursor/caret in the main editor is now a black box to facilitate identification of the currently selected residue. Installation of program updates as well as of auxiliary and emboss programs has been rewritten hopefully making these operations more reliable. The plasmid editor has been updated and now works quite well for construction of simple plasmids. For more demanding construction tasks you should consider using Vector NTI.

Please note that sequence alignment with Clustal W - which has not been working on Windows XP - is updated and now functions as it should - also on XP machines.

The previous version 8.1 can no longer be downloaded nor be updated. To use version 8.2 you will have to reinstall SEQtools and to get a new registration key - unless you have a full license.

06.01.2004 - Various corrections to 8.1 - A new function has been added for batch blast searching at NCBI. The new function is built on the urlapi (QBlast) made available by NCBI and seems to work quite well. The new function circumvents firewall related problems, is quite fast and hopefully do not suffer from the difficulties of the Blastcl3 based old function. The two versions of batch blast search functions will co-exist for a while in seqtools until the QBlast function has been thoroughly tested.

Minor revisions have been made to the Restriction Enzyme functions, to the Retrieve function used to fetch sequence records from Entrez, Sequence Loading.

A simple routine has been added to all internet dependent functions and calls which check if an active connection is available. According to the documentation the api used for this purpose requite that IE 3.0 or newer is installed in the PC. I am not sure if this will create problems for users not using IE. Please send me a mail if this is the case.

04.12.2003 - Various corrections to 8.1 - Several bugs have been corrected and various improvements made in the following functions: EST submission, loading fasta files, fasta definition line editor, blast sequential search, loading and saving project files, loading ms files via the commandline, trimming polyA tails.

05.10.2003 - Various corrections to 8.1 - Several bugs in the EST submission function, Header compose function, Loading of fasta formatted multi-sequence files, Fasta definition line editor. Speed improved by various fine-tuning the code. Please continue reporting bugs in the new version...

Further testing of remaining functions is under way.

03.10.2003 - Various corrections to 8.1 - Several bugs removed and code improved in: 1) Loading of multi-sequence files, 2) Multiple local database blast search and tabular listing of results, 3) Header composition function, 4) Fasta definition line editor, 5) Manual installation of auxiliary and emboss programs, 6) SEQtools installed components listing, 7) Save/Export sequences.

Further testing of remaining functions is under way.

02.10.2003 - Version 8.1 of seqtools released - Handling of information stored in sequence headers has been completely redesigned in version 8.1. The header compose form now gives you complete freedom to display and list any type of information contained in headers. With a few simple checkmarks you can select one or more blast search results plus a number of other information types such as genbank records and medline entries, various types of imported information, local multi-database searches etc.

These options extends to the sequence listing options. From a prioritised list of blast search results you can compose a "virtual" blast section combining data from multiple searches: top priority results are used first. In cases where a result is not available in the top priority search the remaining searches is scanned until a search result is found for the sequence or the end of the result list is reached. In the displayed sequence list it is clearly indicated which blast search provided the information.

In addition large parts of the seqtools code has been optimised to improve speed.

To try seqtools 8.1 you must download the full setup file, accept to remove the old installation before installing version 8.1. Please note that unless you have a full license you need to get a new license to run version 8.1.

Also note that the context sensitive help has not yet been updated to cover these changes. This will happen soon.

29.09.2003 - Save and Export function merged, Sequential Batch Blast function modified - The Save Sequence and the Sequence Export forms have been merged into a single save form which includes the combined functionality of both functions.

The Sequential batch blast function have been extended to include the same options as the old single blast search form - in addition to the sequential search option. The old batch blast function will be removed in the next update.

The new batch blast function has been carefully tested but still lack some internal control of the user interaction. It is for example at the moment possible (although it won't produce any results...) to choose a blastx search as the first search and a blastn search as the second. This will generate a protein in the first search and use this as the query in the subsequent blastn search. This will be fixed in a future revision.

Remember that you are always welcome to contact me in case you encounter problems with the modified (or the old) functions.

I would like to express my sincere thanks to Dr. Anne Lockyer for her persistent work in solving the problems associated with tblastx searches at ncbi. I will report the results of her conversations with the people at NCBI when the case is fully documented.

25.09.2003 - New sequential batch blast function - Sometimes a blast search returns a result which fails to reveal a putative function of the gene returned by the search. To improve this situation a new batch blast function has been added to SEQtools which allows you to perform sequential batch blast searches.

The idea is the following: for each sequence one blast search is performed against one database. The result of this search is evaluated. If a match better than a specified cutoff value is found, the sequence record is retrieved from Entrez. The retrieved sequence is then used in a second search against a different database and/or with different blast program.

Typically this function can be used to annotate unknown short sequences (e.g. 50-mer reporters derived from est sequences) by using the est section of genbank in the first search and nr in the second.

The function is included in the current update but should be used with caution until further testing has been performed.

04.09.2003 - Various improvements - A sequence list field has been added to the main header form making navigation between headers more convenient.

Printing of pipetting templates has been improved by adding an option to print the label on a single line with a slightly larger font.

The erratic bug described in the last newsletter emerging while navigating appears to have been successfully corrected by the latest revision of the relevant code.

A new function has been added to the renaming form. This function reads a tab delimited text file (derived from data contained in a spreadsheet) containing a New name, an Existing name and/or a piece of nucleotide sequence. The function allows you to replace project sequence names by the new names contained in the text file. The function was added for a specific purpose and may not be relevant for most users.

I am planning to include the search options provided by three functions (fuzznuc, fuzzpro and fuzztran) included in the emboss package. So far a download/update routine for the necessary emboss programs has been added to seqtools. The program interfaces to the emboss programs will be added later. So don't worry if you are notified about missing emboss programs.

The www menu on the main sequence editor has been improved by adding an option to edit the URLs list under this menu. Open the General Preferences form and choose the Launch URL tab. The method to edit/add new URLs is the same as for the Launch application tab, i.e., type the URL (no leading "HTTP://") in the left text field and the name you want to see on the menu item in the right field. I have added some default URLs that can be changed to suit your needs.

Remember to install GeneDoc and TreeView on your pc and to include them in the Launch Application menu to get access to the two programs from within seqtools. The ClustalW aligment function (with GCG formatted output) is directly linked to GeneDoc.

Selecting alignment with the ClustalX program with phylip output and performing BootStrap N-J, launches the seqtools Clustal results form with TreeView accessible from the menu (when the ClustalX form is closed). Clicking the TreeView menu item launches this program with the generated php file as input.

Several items in the context sensitive help have been updated. There are still several items that need revision so please be patient.

25.08.2003 - Serious bug corrected - maybe... - As previously described a serious bug was detected causing seqtools to duplicate the sequence part of a record overwriting the sequence of the following sequence in the project without affecting its annotation. When the problem surfaced in March I rewrote the code performing the navigation but apparently did not correct the problem.

I have modified the code a second time in this revision (8.0.781) but as I have been unable to reproduce the problem I can't be absolutely sure that the bug is corrected.

An easy test for the result of this very unpleasant bug is to run Search > Find In Project > Find Identical Sequences. This search function is very fast and will list all project members with identical nucleotide sequences.

If you encounter this problem I would very much appreciate it if you send me a note with as many details as possible. Needless to say that I sincerely regret this bug.

21.08.2003 - Several minor improvements - The batch sequence annotation facilities which auto-annotate sequences based on unattended blast searches and retrieval and parsing of genbank sequence records are some of the very strong features of seqtools. Some users have have asked for options enabling you to manually edit the auto-generated annotation (genbank definition lines). This feature has been added to this compile of seqtools.

The auto-build project from accession/GI numbers has been improved by the addition of download filters enabling you to limit sequences to be retrieved to specified subsections of genbank (ref, pir, emb etc.). The same options are included in the auto-annotate function (Entrez)

Options for saving customised tab-delimited lists including specified columns of the various sequence lists. Such lists are primarily intended for import into spread-sheets.

Previously, when seqtools was launched by double clicking a sequence or a multi-sequence file nothing indicated that loading/unpacking was in progress. Now a small message displays while importing and processing large ms-files.

I am sorry to admit that updating both the web based manual and the context sensitive help is far behind the code reflecting that it is a lot more fun to design new features that to describe them. Please write when/if you fail to locate a particular function. Updating the help/documentation is slowly crawling up the priority ladder. It may, however, take some time before it gets to the top...

Some of you may have noted that setup and update files on the download page usually include the same compile: I have recently gotten a fast adsl connection to replace my old modem, so uploading the setup file is no longer a 3-hour process.

07.08.2003 - Find antisense sequences in project - A new function has been to the collection of search functions. The new function help you find alignments after a blastn search where the project sequence is complementary to a genbank sequence. This is critical when you plan to use project sequences for microarray oligonucleotide design. This is primarily relevant when EST sequences are used for primer design.

As usual several minor bugs, including one in the fasta sequence export option, have been corrected. Also as usual the documentation is lagging somewhat behind.

I also wish to remind you to contact me when you encounter strange program behaviour and bugs or lack a function in the collection for your particular purpose.

24.06.2003 - Sequence names - The options for changing/modifying sequence names have been reorganised and extended. It is now possible to perform changes to sequence names on selected sequences in the sequence list. A new option has been included enabling you to use accession numbers, long sequence names, gene or locus names. In cases where these names are not available the original sequence names are used.

07.06.2003 - Clustering - As a result of extending the number of allowed characters in sequence names to 16 in seqtools, the reports generated by the clustering function produced scrambled output - or no output at all. This has been corrected in the current revision. The three clustering methods include clustering as created by the NCBI clustering program, clustering based on blast score values and clustering based on shared identical sequence regions of specified length and with specified number of mismatches.

The clusters as listed in the report are linked to sequence header data and to clustal alignment for further analysis of individual clusters.

03.06.2003 - Duplicate sequences - New functions have been added to identify and close duplicate sequences in a project. The form includes three different options for identifying redundant sequences:

1) searches for identical sequences and places the two members of a pair of identical sequences in separate lists,

2) compares alignments generated by a local blast search against a local database. This facility enables you to quickly identify whether sequences contained in the project are present in the collection of sequences from which the local database is generated.

3) finally, a facility is included which compares all sequences contained in the project against each other. The shorter member of a matching pair is placed in one list, the longer one in the other list.

The function allows you to select and remove sequences in either list from the project and to save tab delimited lists containing one, the other or both lists.

About emboss - The comprehensive collection of programs for sequence analysis, EMBOSS, has recently been compiled for win32. These command line programs can very easily be integrated into seqtools. In case you need functions included in the emboss package, please let me know.

21.05.2003 - Minor bug fixes - Bug fixes in formatting/verifying filenames. Small changes in microtiter plate index file function. Implementation of EMBOSS functions under way. Change in NCBI script for retrieval of GI numbers and sequence records implemented. Hopefully, the batch retrieval of sequence records from Entrez now works again.

I failed to realise that the NCBI staff made a minor but quite important correction to the parameters for converting accession number to gi numbers. In this context I would like to encourage seqtools users to contact me whenever functions fail to produce the expected outcome. It is difficult for me to monitor all functions regularly to detect problems like the change of Entrez parameters.

About emboss - The comprehensive collection of programs for sequence analysis, EMBOSS, has recently been compiled for win32. These command line programs can very easily be integrated into seqtools. In case you need functions included in the emboss package, please let me know.

18.05.2003 - Blast description line formatting - When working with different projects the criteria for formatting information included in sequence lists often different. Before is was necessary to compose the criteria used for trimming/selecting each time a different project was loaded. An option is added to this revision enabling you to save and load formatting schemes used in the sequence list settings.

The use of logical operators in the header search function has been improved and now hopefully interprets AND, OR and NOT correctly.

Minor additions have been made to the function handling the composition of new projects from blast search results.

A bug in launching a new instance of SEQtools containing highlighted sequences in various forms has been fixed.

09.05.2003 - New function added - It is sometimes convenient to be able to transfer sequence information (selected subsections of the sequence header) from one project to another. A new function has been added in revision 8.0.749 which allows you to import selected subsections from an existing project saved as a SEQtools formatted multi-sequence file into the headers of sequence contained in a new project.

Before import is enabled the source multi-sequence file and the current project are examined to identify available information in the source file and to compare the sequence names (both short and long sequence names) of the source and project files. If the source file contain subsections selectable for import and if the source file and the project contain sequences with identical short and long names, import is enabled letting you select among the available subsections in the source file. You find the function under the retrieve menu of the main editor and under multi-sequence functions, tools menu.

A bug in the header compose function making deletion of header subsections a somewhat unpredictable affair. The problem is corrected in this revision.

05.05.2003 - Minor adjustments - Various minor adjustments have been made in this revision.

30.04.2003 - Description line formatting options extended - The description line formatting / editing options for sequence listing has been modified to improve flexibility. This has made it necessary to change the format of the ini-file (more space was required to hold the larger number of items in the words lists). When you start seqtools after this upgrade you will be asked to create a new ini-file for your project. This also implies that you must re-enter your preferences, sorry.

About emboss - The comprehensive collection of programs for sequence analysis, EMBOSS, has recently been compiled for win32. These command line programs can very easily be integrated into seqtools. In case you need functions included in the emboss package, please let me know.

25.04.2003 - Context sensitive help updated - A few minor revisions are included in this revision - and the context sensitive help has been updated to cover the latest additions to SEQtools (several microarray related functions). There are, however, still several older functions where the context sensitive help provided does not exactly reflect the content of forms.

20.04.2003 - SEQtools folder structure, removing sequences from project - The sub-folder structure of program and DOS folders has been changed to make subdivision of various data types more logical.

The function scanning the project for identical files has been improved to make it easier to detect and remove duplicates (also if they have different sequence names).

Program installation is now handled by the microsoft software installer (*.msi) which makes it simpler - for me - to maintain the installation program.

14.04.2003 - Bug in create local db's, updated FastA def. line editor - A bug preventing the proper operation of the function for creating local databases has been fixed.

The fasta definition line editor has been re-designed to improve its functionality.

The microtiter plate indexing utility has been improved to support printing a 8 x 12 pipetting template facilitating aliquoting samples.

02.04.2003 - Retrieving data from Entrez - The utilities for data retrieval from Entrez (sequence records, medline records and GI numbers) have been rewritten using the new cgi scripts made available by NCBI. Like the blast function the Entrez retrieval functions have been made more robust resuming the retrieval if the first attempt failed. The error-recovery handles server problems (time-out, no-response) as well as program errors.

Retrieved genbank sequence records can now be used to list sequences with the same formatting options as previously available for default blast description lines.

22.03.2003 - Header search, Filtering of sequence lists, FastA definition lines, Help file - The function for searching sequence annotation has been improved: You can now use boolean operators and have the matches coloured in the displayed header. As before the first step in searching the annotation is to compose the displayed header by selecting subsections to be included.

The sequence list function now has a filtering option working on all displayed formats. This feature simply hides entries if the first 50 characters are identical and only includes the first occurrence in the list. This feature is especially useful when handling EST projects.

Inserting blanks when indexing microtiterplates has been made a bit more sophisticated: you can now create a pattern/template describing which positions to leave empty. The pattern will remain in effect until it is changed or the form is closed.

As some of you may have detected you can compose your own FastA definition lines with the SEQtools FastA definition line editor (to be used for example when creating local databases). Previously user generated definition lines replaced existing definition lines imported when the sequences were loaded. This is no longer the the case, imported and user generated FastA definitions are now stored separately. Remember that information can be extracted from the displayed header and included in FastA definition lines with or without flanking user comments. Altogether this makes it possible to compose quite informative annotation for local databases.

The context sensitive help has been updated to include some of the latest changes/additions to seqtools. This does, however, not imply that the help is complete yet.

hope you enjoy my two friends above...

20.03.2003 - Alignment with ClustalW - A small modification somewhere else in the code unfortunately made it impossible to select files for aligning with Clustal. This problem is corrected in this update.

Handling of imported and user created FastA definition lines has been improved. It is now possible to compose definition lines with theFastA definition line editor (to be used for example when creating local databases) without at the same time erasing imported FastA definition lines. The Compose Header form now contains separate subsections for FastA definitions.

17.03.2003 - Duplicate filenames, remove sequences from project - Handling of duplicate filenames improved: all routines dealing with verification of file/sequence names in projects have been rewritten and hopefully now works as intended. The function removing sequences from a project has been rewritten to improve speed. The microtiterplate indexing function has been improved to facilitate indexing multiple plates from sequences contained in a project.

By mistake an incorrect version 8.0.723 was uploaded. Users who accepted to upgrade to this version should manually download this version and replace version 8.0.723. The procedure for manually updating is described on thedownload page.

06.03.2003 - Serious bug detected - I recently detected while processing a large set of sequences to be included in a microarray, that rapid navigating from a long (10-50,000 bp) sequence to a short (500-1,000) EST sequence in rare cases resulted in the replacement of the short nucleotide sequence by the long. The annotation remained "correct" but the associated sequence no longer corresponded to the annotation.

Careful inspection of the relevant code did not provide any hints as to the source of this problem. The update routine which stores editorial changes to the displayed sequence has now been rewritten and so far I have been unable to provoke this behaviour after changing the code.

I have not had any user comments about this problem and hope this implies that I am the only one having experienced the problem. Needless to say that I sincerely regret this bug.

If you open more than one instance of SEQtools, you can now color code each instance, Preferences>Application Color Coding, to make it easier to distinguish which of the open forms that belong to a certain instance. Without this it can be rather confusing when several instances of SEQtools are running simultaneously.

25.02.2003 - Fixes minor bugs - A property value incorrectly set in the code prevented selection of multi-lines in the new sequence listing form. This problem is corrected now.

A small improvement to batch blasting is a new feature included in the ncbi blast form. If the sequence listing form is displayed and one or more lines are highlighted, a checkbox on the ncbi blast form allows you perform the blast search only for the selected sequences.

If you are experiencing problems with the auto-update function I would very much like to hear about it.

23.02.2003 - Auto-installation of auxiliary programs - SEQtools uses several third-party programs, contained in auxiliary.exe, for various functions, the most important ones being the blast programs from NCBI for sequence comparison and ClustalW written by Toby Gibson, Des Higgins and Julie Thompson for sequence alignment. These programs are not included in the core download but must be installed after setting up SEQtools.

Despite my carefully worded explanations, this operation has been difficult for many users. In this update I have made this process a bit easier: When SEQtools starts it checks if the auxiliary programs are correctly installed. If not it offers to download and install the programs before it continue loading. When done the SEQtools configuration form is displayed so you can check if everything is properly installed.

Note that the size of the file is 6.2 Mb and that there is (not yet...) any progress bar, so be patient while downloading is in progress.

21.02.2003 - Auto-notification when a new SEQtools update is available - After several attempts to implement his facility the automatic notification routine informing users about new updates now seems to work (at least when I test it...).

Every time you start SEQtools, the program checks the download web page to see if a new update is available. If a newer version of SEQtools is available you can choose to download it. If you have an extended demo license the message box also informs you whether the new update is covered by your license or if you need to repeat your registration. Users with a full license have unlimited access to updates.

After downloading the update is automatically installed when you close the last instance of SEQtools.

20.02.2003 - Auto-retrieval of sequences from Entrez - As a result of my current work designing new microarrays a number of routines have been added to SEQtools. The functions enables you to "convert" various types of accession number lists to SEQtools projects by auto-retrieving the nucleotide sequences from Entrez.

Functions have also been added to transform a project consisting of for example human sequences to a new project containing the corresponding rat or mouse orthologs.

19.02.2003 - Improved sequence listing - The design of the sequence listing function in SEQtools has long been very unsatisfactory. This rather critical part of the program has been improved by replacing the old listbox control with the more flexible listview control.

The listview control allows you to sort by selected columns simply by clicking the column title field. Unfortunately the control is unable to sort by numerical values. This implies that you cannot sort by expect values (sequence length, however, works fine as SEQtools converts these numbers to fixed length strings).

The options for sequence listing have been changed to include, in addition to standard sequence information, description lines provided by blast search results, Fasta definition lines (available when the project is created from a fasta multi-sequence file and information extracted from Genbank annotation (either genuine Genbank headers or abbreviated Genbank information created by the SEQtools export function and the Create Ortholog project function).

15.02.2003 - Sequence listing form redesigned - It is now possible to sort the sequence list with an individual column as sort key.

Pressing F5 in the main editor now auto-creates a local database including all sequences in the project and performs a blast search with the current sequence. The matches are listed in the sequence index making it very easy to find and retrieve similar sequences in the project.

Integration of Entrez and LocusLink improved in sequence headers (without hyperlinks). Just click an accession number in the header text and click the "Entrez" command button to display the Entrez/Locuslink information.

30.01.2003 - Blast at NCBI - Bug in the NCBI blast function corrected.

03.01.2003 - SAGE analysis - Loading and analysis of SAGE profiles optimised.

26.12.2002 - Bug corrections - Loading of multi-sequence files substantially improved. A few bugs in the two functions for editing file names corrected. Several bugs in multi-database local blast search and result display corrected. The collection of multi-sequence tools optimised (but not yet tested very carefully...). A bug in the function converting enzyme file from GCG format to SEQtools corrected.

19.12.2002 - Bug corrections - Several bugs have been corrected in the latest update (8.0.671): the clustering functions now also handle protein sequences, the re-naming functions now accept sequence names up to 16 characters long, plus a number of smaller improvements most of then related to the increase in maximum length of file names (from 8.3 to 12.3/16). Speed has been improved by code optimisation in most functions (at least according to my code profiler) which also informs me that seqtools source code contains some 120,000 lines of which 66,000 are executable. No wonder that I sometimes feel a bit tired...

07.12.2002 - SEQtools published... - I wish to thank the large number of users for their kind remarks and encouraging comments to seqtools. There are probably still a number of corrections and adjustments to be made before the software package is fully functional. I hope you will continue to report any misbehaviour of the program as well as send me your suggestions for new functions. Remember to check the homepage regularly for updates.

22.10.2002 - SEQtools published - Rewritten from scratch SEQtools now appears for the first time as version 8.0.