MEME Suite System Release Notes

MEME version 4.8.1 -- Feburary 6, 2012

  • New Features and Enhancements
    • CentriMo
      • Added method to create custom Encapsulated Postscript files of the motif probability graph.
    • MEME-ChIP
      • Added SpaMo to the list of programs that MEME-ChIP runs.
  • Bug fixes
    • Fixed a typo which was breaking the "Upload your own database" feature on Tomtom.

MEME version 4.8.0 -- January 16, 2012

  • New Features and Enhancements
    • CentriMo
      • Added HTML output with lots of interactive features allowing selecting, sorting and filtering of which motifs to display and plot the site-probability curves for.
      • Added CentriMo web submission form.
      • CentriMo now runs as part of MEME-ChIP.
      • Can now use motif matches to the given strand only if the -norc switch is given. Useful with RNA sequences.
      • Removed centrimo-plots program as the new HTML output replaces and improves on its purpose.
    • DREME
      • Can now look for motifs on the given strand only if the -norc switch is given. This can be useful with RNA sequences.
      • Using given strand only now available as an option on the web submission form.
    • MCAST
      • Added support for scanning genome scale sequences.
      • Added q-values for match scores.
      • Revised HTML output to include block diagrams of matches.
    • SpaMo
      • SpaMo now normalizes the p-values reported for particular motif spacings (bins) only for the number of spacings (bins) tested. Previously, SpaMo reported p-values normalized for both the number of spacings and the number of motifs tested.
      • SpaMo now prints E-values for secondary motifs, which is the lowest p-value of any spacing (bin) times the number of secondary motifs.
      • Added -evalue switch to specify largest E-value allowed for secondary motifs.
    • Other
      • Improved memory efficiency for reading FASTA sequences in
  • Bug fixes
    • Fixed segmentation fault in FIMO when option --bgfile motif-file is used.
    • Fixed png generation code for custom motif logos.

MEME version 4.7.0 -- September 22, 2011

  • New Features and Enhancements
    • DREME can be accessed as a webservice directly which allows use of the discriminative motif discovery.
    • DREME now produces XML and HTML outputs as well as text. These new outputs are accepted where MEME motifs are accepted.
    • New command-line program CentriMo and the associated CentriMo Plots for determining which motifs are centrally enriched in ChIP-seq output.
    • Completely rewrote motif loading code. The main user visible difference is that the minimal MEME format is a lot more forgiving of spacing and MAST can now accept the minimal MEME format without any PSSM section.
    • Output fromm all tools nows lists prefered citation.
    • Updated and re-formated list of publications.
  • Bug fixes
    • Occasionally MEME would produce HTML files which could not be used as a motif input to other programs. The new motif parser should be able to read these files and hence fix this problem.
    • SpaMo background is now calculated from the sequences as stated in the paper.
    • The website menu used to use the full URL which unfortunately resulted in people bookmarking specific versions of the MEME suite rather than the URL for the latest version. The website menu has been corrected to use only relative URLs.
    • Links should now be created to the error messages when a webservice fails. This used to work but some setting changed the way redirects are buffered.
    • Added compiler/linker directives to set stack size for Cygwin to 9MB. Cygwin defaults to a 2MB stack, which is too small.
    • MAST was not handling the -remcorr switch. When a motif was removed because it was too correlated, the indices of the other motifs could be incorrect. Fixed.
  • Known issues
    • DREME does not yet support single strand scanning. It is planned but wasn't ready for this release.
    • Uploading extremely large sequence sets to MEME-ChIP can cause a HTTP timeout and/or failure of the CGI script.

MEME version 4.6.1 -- March 21, 2011

  • New Features and Enhancements
    • MEME-ChIP command line tool --
      MEME-ChIP was released in 4.6.0 as a web service but it did not have a command line equalivent. The webservice has been rewritten as a command line tool.
    • TOMTOM --
      The default TOMTOM scoring algorithm now considers all columns in the query motif, not just the aligned columns. This reduces the number of spurious alignments including uninformative columns. The older scoring system is available via the -incomplete-scores option.
  • Bug fixes
    • MEME-ChIP --
      • MEME-ChIP has been rewritten to avoid infinite loops that could occur when the webservice was passed parameters including colons.
      • The webservice now updates its output with the completion of each sub-program so if it is teminated by Opal for running too long the already completed output is still avaliable.
      • Error reporting has been improved with the result that MEME-ChIP should not be able to fail mysteriously as it has on occasion been doing.
    • FIMO
      • The --norc option was being ignored. This has been fixed.
    • Motiph
      • The --seed option has been added to set the seed for the random number generator.
  • Known issues
    • MEME-ChIP when given very large sequence sets will not complete before the Opal webservice time limit.
    • Uploading extremely large sequence sets to MEME-ChIP can cause a HTTP timeout and/or failure of the CGI script.

MEME version 4.6.0 -- December 24, 2010

  • New Features and Enhancements
    • MEME-ChIP web service --
      • Motif Analysis of Large DNA Datasets
        The MEME-ChIP service allows submission of a FASTA file of DNA sequences of unlimited size and with minimal parameter settings (only the traditional MEME web form settings) and runs it through a variety of motif finding and analysis tools.
      • dreme --
        added dreme and integrated it into the MEME-ChIP into web service.
      • AME --
        added ame and integrated it into the MEME-ChIP into web service. This is a cut-down version of the ame previously described, only supporting one mode of motif enrichment analysis.
      • fasta-dinucleotide-shuffle --
        added scripts/fasta-dinucleotide-shuffle and integrated it into the MEME-ChIP into web service. Thanks to Peter Clote for allowing use of his dinucleotide shuffle code.
      • fasta-center --
        added fasta-center and integrated it into the MEME-ChIP into web service.
    • SpaMo beta release --
      The web interface and documentation is currently rather limited but in the interests of getting this out for review this is a preview edition of the Spaced Motif analysis tool.
    • FIMO --
      The FIMO HTML output now explicitly labels the strand, rather than indicating the reverse strand by having start > than stop. The HTML output also now includes the matched sequence. The GFF output is now GFF 3.

MEME version 4.5.0 -- September 6, 2010

  • New Features and Enhancements
    • Tomtom --
      • Search multiple databases
        Tomtom can now search more than one motif database. To keep the ability to have web links for each motif they were moved into the MEME motif format itself. See the documentation.
      • Improvements in submission form
        Tomtom's submission form has many new enhancements. Motifs can now be specified in 3 formats: IUPAC motifs, frequency matrices, and MEME format files. Go try it.
      • Improvements in output format
        Tomtom now outputs an XML file which it converts into the HTML output. The HTML output can dynamically draw the motif logos when the images aren't available on web browsers supporting HTML5. Additional improvements include a summary table at the top giving direct links to the top 20 matches for each motif. See the sample.
    • FIMO --
      The output for FIMO now includes a wiggle track format file.
    • MEME file format --
      The MEME file format now includes an optional URL for each motif. This is used to create the links in the Tomtom output. All the XXXX2meme scripts have been retrofitted to take this URL parameter.
    • ama-qvalues --
      New script. See the documentation.
    • fasta-subsample --
      New script for extracting a random sampling of the sequences in a FASTA file. This is especially useful for ChIP-seq peak datasets to be input to MEME. Using this script, a FASTA file containing a subset of the ChIP-seq peak sequences (or any other FASTA file) can be created. The total number of sequences should be less than 1000 (preferably less than 500), and the total sequence length should be less than a few hundred thousand. MEME typically takes about 20 minutes per motif with files of 100,000 characters (DNA, both strands, ZOOPS model), and scales quadratically in the total sequence length (so a file of 200,000 characters will take four times as long.) This new script can also output the remaining sequences (in a seperate file) for use in cross-validation. See the documentation.
    • meme2meme --
      New script. See the documentation.
    • rna2meme --
      New script. See the documentation.

MEME version 4.4.0 -- April 23, 2010

  • New Features and Enhancements
    • MCAST --
      MCAST now been integrated into the MEME Suite web applications. MCAST is a tool for searching for statistically significant clusters of motifs in DNA sequences. MCAST was previously supported on a separate web site.
    • MEME
      • Discriminative motif discovery
        The MEME web server now supports discriminative motif discovery. The user may now provide a set of "negative sequences" in addition to the normal "positive sequences", and MEME will look for motifs that are overrepresented in the positive sequences relative to the negative sequences.
      • Improvements in output format
        MEME's html output (sample) has been reorganised to make important information more accessible as well as to make it aesthetically pleasing. Some of the improvements include:
        • A preview of the motifs found, including their reverse complement for DNA, in miniature at the top of the page.
        • Ability to view full sized reverse-complemented DNA logos.
        • Highlighting of all output for a sequence with a single click.
        • New block diagrams that graphically show the strength (and DNA strand) of motifs, and that scale with the size of the browser window.
        • Support for the new web standard HTML5. This allows the motif logos to be viewed without the images on supporting web browsers (Firefox, Chrome, Safari). Unfortunately Opera has inadequate text support for canvas (the new HTML5 element required for this feat) and Internet Explorer currently doesn't support canvas at all. It is anticipated that Internet Explorer 9 will support this feature.
    • MAST --
      • Improvements in output format
      • MAST's html output (sample) has been reorganised to make important information more accessible as well as to make it aesthetically pleasing. Some of the improvements include:
        • Everything is now in one place - no longer do you have to search to find the associated annotated sequence for that high scoring sequence, as with one click the hidden information is displayed directly below. Additionally if DNA strands are scored separately then the sequence only appears in the list once based on the better scoring strand.
        • New block diagrams - incorporating all the improvements in the new MEME block diagrams such as graphical display of motif strength, scaling with the browser window and the DNA strand. As an added bonus separately scored strands are visible on the same diagram.
        • Interactive annotated sequence - a new range selector allows you to view the sequence around the motif hits of your choice, directly relatable to the block diagram. For separate DNA strand scoring only one strand is visible at a time.
        • Now in colour - the annotated sequence for motif hits now have letters colour coded using the same colour scheme as MEME motifs.
      • MAST now outputs xml in a similar fashion to MEME. This has resulted in differences to how you would invoke MAST from the command line. Additionally it no longer has the script wrapper.
      • MAST now uses an xml style sheet to create its html output.
    • mast2txt --
      Created program mast2txt for outputting the old text format for MAST. Note that there are some problems with backwards compatibility in separate scoring mode. The use of the text format is no longer recommended.
    • psp-gen --
      added psp-gen and integrated PSP generation into web service.
    • MHMM --
      Fixed bug in --order option. The option was handling motifs with strand identifiers incorrectly.
    • TOMTOM --
      Added -no-ssc switch to fix problem with motifs with few sites having empty logos; this switch is now the default for the TOMTOM website.
    • GOMO --
      Improved gomo html format. Added highlighting of the most specific GO terms in the html output. This is activated by the switch --dag which expects a GO DAG file.
    • gomo_highlight --
      Added a program to post process gomo xml files to identify the go terms which are implied by other more specific go terms to allow the specific terms to be highlighted in the html output. This is used by gomo when the --dag switch is specified.
    • obo2dag --
      Added a tool for generating a GO DAG file from a Gene Ontology OBO file.
    • GO DAG format --
      Documented the new GO DAG file format and improved GOMO documentation as well as the command line overview.
    • GLAM2 --
      The output now prints a regular expression in addition to the logo for each motif found.
    • MEME motif format --
      Fixed errors in motif format description doc/meme-format.html and example files.
    • meme-io --
      Updated meme-io to accept the new MEME html format.

MEME version 4.3.0 -- September 26, 2009

  • GOMO --
    Added comparative genomics to GOMO. GOMO can now use multiple, comparative genomes in making its predictions. This substantially increases the sensitivity of predictions of roles for DNA binding motifs. The GOMO web-service now uses GC-compensated AMA p-values.
  • AMA --
    P-values are no longer printed by default. A switch has been added to cause them to be printed.
    AMA can now compute GC-content compensated p-values. The GC-content of an individual sequence is used in estimating the p-value of its AMA score.

MEME version 4.2.0 -- June 22, 2009

  • MEME --
    MEME now can use position-specific priors (PSP) to improve motif discovery. The PSP is specified using the -psp pspfile switch to MEME.
  • GOMO --
    Added the ability to choose either maximum or average binding affinity as the scoring function.
    Fixed a bug--was using maximum affinity by default, contrary to the documentation.
  • AMA --
    AMA now computes p-value of each sequence based on a 0-order background model for the dataset.
    Improved speed and memory efficiency; was re-reading sequences and storing them in memory.
    Fixed bug--was using max-odds by default and the --scoring switch was being ignored.
  • MAST -- website now supports protein databases even when DNA database contains long sequences.
  • Installation --
    Now all programs in the MEME suite are built. The --enable-both configure switch is now obsolete.
    Simplified the build process by removing support for separately building the tools formerly included in Meta-MEME.
    Update on-line documentation.
    Simplified "quick install" instructions are provided in INSTALL and in doc/meme-install.html.

MEME version 4.1.1

  • Improved documentation.
  • Increased default number of iterations for GLAM2.
  • Improved error handling for sequence data.
  • Added links to server activity summary.
  • Fixed bug in install process when --enable-both not used.

MEME version 4.1.0

  • New tool: GOMO -- Finding Genome Ontology terms associated with DNA binding motifs.
  • Added FIMO and GOMO buttons to MEME output to make it easy to use discovered motifs to search sequence databases and to identify the role of DNA binding motifs.
  • Fixed placement of sidebar menu in Internet Explorer for output of CGI scripts.
  • Fixed support for OpenMPI in MEME.
  • Improved the efficiency of FIMO.
  • Made the -text option to FIMO produce results on-the-fly, without sorting or q-values. This mode allows very large databases to be searched, without storing intermediate results.
  • Modified the command line syntax for gendb, for consistency with the rest of the suite.

MEME version 4.0.0 -- June 13, 2008

  • New Features and Enhancements
    • MEME and Meta-MEME are now integrated as "The MEME Suite".
    • New web-based tools:
      • GLAM2 and GLAM2SCAN--discovery and scanning of gapped motifs
      • Tomtom--comparing motifs
      • FIMO--basic motif scanning (complements MAST)
    • Enhancements of existing tools:
      • Sequence LOGOS now enhance MEME output.
      • Compare new motifs to databases of motifs via Tomtom by clicking an easy-to-use button on MEME and GLAM2 output.
      • Improved local search (branching search) in MEME (command line version only).
      • Hundreds of new genomic and upstream sequence databases are now supported for scanning by MAST, FIMO or GLAM2SCAN.
      • Tomtom supports four motif databases: JASPAR, Transfac, SCPD and MacIsaac Yeast.
    • The MEME Suite tools now have web servers powered by OPAL, and return their results via the web, rather than by email.
    • A single web server portal simplifies different types of motif scanning tasks

MEME version 3.5.7 -- December 14, 2007

  • New Features and Enhancements
    • Parallel MEME is compiled when no scheduler is found but MPICH is detected.
    • Opal integration with meme.
    • Added --enable-webservice=<path_to_build-opal.xml> configure flag. Used to automate deployment of MEME inside an Opal installation as a web service.
    • Added --enable-web=<opal_url> configure flag. Used to enable building of the MEME web portal. A URL for an Opal-exposed MEME web service must be entered.

MEME version 3.5.5 -- May 31, 2007

  • New Features and Enhancements
    • MAST -hit_list can now handle very long sequences and prints ALL hits regardless of the sequence E-value cutoff. It splits them into "swaths". The -hit_list switch now overrides the -seqp switch, and hits are NOT sorted by the E-value of the sequence.

MEME version 3.5.4 -- September 21, 2006

  • New Features and Enhancements
    • Changed the MAST -hit_list behavior to print ONLY a list of the non-overlapping significant motif matches in each sequence in a machine-readable format. (More complete documentation is given in the MAST "usage" message. Type "mast" on the UNIX command line to see the usage message.) An example of the (complete) output of MAST using the -hit_list switch is:
                    # All non-overlapping hits in all sequences.
                    # sequence_name motif hit_start hit_end score hit_p-value
                    ce1cg -2 8 22  1459.90 1.67e-06
                    ara +2 2 16  1661.18 5.04e-08
                    bglr1 +2 1 15  1274.97 1.42e-05
                    cya -2 19 33  1101.37 6.64e-05
                    gale +2 5 19  1076.21 8.11e-05
                    ilv -2 6 20  1098.85 6.78e-05
                    malk +2 37 51  1085.02 7.56e-05
                    ompa +2 5 19  1583.18 2.43e-07
                    # mast tests/meme.crp0.oops tests/crp0.s -stdout -hit_list -m 2
    • New "regular expression" section added to each MEME motif.
  • Bug fixes
    • Fixed printing of command name in MAST results.
    • Fixed the handling of the "View FASTA", "View RAW" and "View motif summary" buttons on the MEME form.
    • Improved documentation of the "Motif Summary" section of MEME output.
    • Fixed bug that prevented having multiple MEME and MAST servers running on a single host by different users. Each server is distinguished by the socket number it listens on.
  • Known Issues
    • The "MetaMEME" button on the top of MEME results page is still not functional. Users are advised to go to the MetaMEME web site and submit manually at this time.
    • MAST: Uploaded file names containing quotation marks may cause the search to fail.
    • MAST: Uploaded sequences in DOS/Windows format may cause the search to fail.

MEME version 3.5.3 -- April 20, 2006

  • New Features and Enhancements
    • The capability to search three JASPAR databases of DNA motifs for matches to a MEME-discovered motif has been added to the MEME output form. Pressing the 'COMPARE PSPM' button on the results form of a MEME search of DNA sequences will submit the selected motif PSPM to the JASPAR 'Compare' server. This will allow you to determine if your motif is similar to any known motifs contained in one of the three JASPAR motif databases.
  • Bug fixes
    • The "MAST" button on the top of MEME results page now automatically loads the motifs from the MEME results for MAST search. A user only needs to choose the database (uploaded or server side) to search against.

MEME version 3.5.2 -- March 1, 2006

  • New Features and Enhancements
    • Newly designed web interface using javascript and css styling.
    • Add support for Macos X server startup.
    • Add meme-install.html with detailed explanation of installation procedures.
  • Bug fixes
    • In "configure" remove check dependency on parallel condition when creating server executables.
    • Path to images/ in cgi scripts.
    • Description syntax in linux startup script meme.linux.
    • Provide fix for a cygwin's new line representation when searching strings.
    • Bug in e_step (oops.c)
    • Bug caused by typo "MEME_BINbin"
    • Remove use of min() in
    • Remove obsolete argument sample_prob from subseq7()
    • Remove OS-dependent declarations of accept(), bind(), connect(), socket(), listen(), htons(). They are defined in system header files.

MEME version 3.5.1 -- December 15, 2005

  • Bug fixes
    • Revert binary and scripts names to what they used to be in version prior to 3.5.0. The binary executables have an extension ".bin", and the shell scripts have their extension ".csh" removed. This allows to keep compatibility with earlier versions.
    • Change startup scripts to start servers as a user specified during configuration. If none were specified, then the uid of the user who runs configure will be used.
    • Always install files in etc/.
    • Always install in lib/perl/.
    • Move meme-explanation.html from web/html/ to etc/. This allows to run scripts that do conversion to html when web site is not installed.
  • Enhancements
    • Added support for Macos X (serial version only, without server)
    • Users can now compare DNA motifs to the motifs in the JASPAR database of transcription factor binding site motifs. A button is provided in the PSPM section of each DNA motif on the MEME output form for this purpose. Clicking on the "Compare PSPM x to known motifs in JASPAR database" sends the motif to the JASPAR website, which returns any similar motifs that are found.
    • Add meme_config.csh file that is used automatically by csh scripts to set needed environment variables.
    • Move job.out file (created when SGE scheduler is used) to LOGS/
    • Add '--with-serial' option to '' for compiling only serial version
    • Disable installation of web by default. To install use --with-web configure option.
    • Lower prerequisite version for autoconf to 2.53, and add missing functions in m4/. Add check for autoconf version in bootstrap, and use m4 functions when needed.
    • Change option '--with-mpi' to '--with-mpidir'

MEME version 3.5.0 -- September 15, 2005

This is the 12th year of MEME development. The current anniversary maintenance release for MEME contains a number of improvements to the installation and configuration of MEME/MAST servers and clients. Thanks to Nadya Williams of SDSC/NBCR for this work.

  • Refactoring of installation and configuration
    • Restructure source directory tree to ease configuration.
    • Highly simplified and optimized installation and configuration for MEME and MAST servers and clients using GNU autoconf/automake tools.
    • Multiple configuration options available via arguments to "configure".
    • Default installation of MEME, MAST server and clients, as well as the web site.
    • Separate installation of server, client or web site if desired.
    • Create a small test suite for 'make test'.
    • Implement procedures for standard 'configure', 'make', 'make test', and 'make install'.
    • Create scripts to run MEME/MAST servers as services, and automatically restarted them when physical host server reboots. Currently, Linux and SunOS are supported.
    • Simplified start of the servers with 'start-mast' and 'start-meme' scripts.
    • Encapsulate site-specific variables in for perl scripts and in meme_config for sh scripts.
    • Add csh and sh configuration scripts to set working environment.
    • Availability as a Rocks Roll for integration with Rocks based cluster system.
  • Enhancements
    • Sent MEME/MAST search results via an email as attachments, rather than inlined in the message body. This allows the MEME output to be saved properly for subsequent use with MetaMEME. It also reduces the chance of the output being modified by email filters.
    • Check email addresses for entry accuracy using javascript. Thanks to Chris Misleh of NBCR for this feature.
    • Check email addresses for valid domain names.
    • Create new 'runtests' script to run all test suites. Simplify parsing of the output.
    • Update FAQ(s) for MEME and MAST.
    • Separate web-related files by type (html, cgi or image) and create a dir structure to support it. This simplifies web maintenance.
    • Add examples of database and motif sequences files.
  • Known Issues
    • The new meme and mast are binary executables, not shell scripts as they used to be. When run with the appropriate arguments, they produce text (non-html) output. To get the "old" behavior of meme and mast with the html output, use meme.csh and mast.csh (without -text option). The following table gives new vs. old usage comparison.
                                 old                                      new
                        meme myfile -dna -text > meme.res       meme myfile -dna > meme.res         # output will be a text file
                        meme myfile -dna > meme.res             meme.csh myfile -dna > meme.res     # output will be an html file
                        mast meme.res                           mast.csh meme.res                   # output will be an html file
  • Support Team Members Wilfred W. Li, Ph.D., Nadya Williams, M.S., Chris Misleh, and Timothy Bailey, Ph.D.

MEME version 3.0.14 -- July, 2005

  • Bug Fixes
    • an array out of bound error when MEME is run in parallel mode using certain DNA sequences with the "look for palindromes only" option enabled. Thanks to Tim Kaiser of SDSC/NBCR for the bug fix.
  • User Support
    • all user questions are directed to our support team

MEME version 3.0 -- December, 2000

  • MEME enhancements
    • Hypertext output: MEME now reports its output in hypertext (HTML) format with appropriate internal hot links among the results and the self-contained documentation on how to interpret them. The alignments are color-coded by nucleic acid or amino acid category.
    • Direct MAST search: MEME motifs can now be used to search sequence databases using MAST simply by clicking on a button on the MEME HTML output document in an HTML-capable email reader or browser.
    • Direct BLOCKS search: MEME motifs can now be submitted to the BLOCKS multiple alignment processor by clicking on a button on the MEME HTML output document in an HTML-capable email reader or browser. This allows MEME motifs to be converted to LOGOS or trees, and to be used to search other databases of motifs.
    • New objective function: MEME searches for motifs that optimize the statistical significance of the log likelihood ratio of the occurrences of the motif.
    • E-values: MEME computes and reports the statistical significance of motifs as the E-value of the log likelihood ratio. This provides an objective measure of how likely the motif is to be biologically significant.
    • E-value stopping criterion: MEME will now stop when a motif whose E-value is above a user-given threshold is found. This guarantees that only motifs with a given E-value or better will be present in the output.
    • Handling reverse complement DNA strands: MEME now handles reverse complement DNA strands correctly with all model types: OOPS, ZOOPS and TCM.
    • Handling of ambiguous characters: MEME now handles ambiguous DNA and protein letters by converting them to the character "X". MEME treats the "X" character as "unknown", and correctly computes the probabilities of motif occurrences containing it.
    • Higher-order background models: MEME now allows the user to specify a Markov model of arbitrary order via a file of tuple frequencies. This appears to improve the ability to discriminate between biologically interesting DNA motifs and motifs that are artifacts of the higher order statistics of DNA sequence.
    • Multiple-alignment-based motif trimming: MEME defines a motif as a set of subsequences that can be correctly aligned without gaps. MEME can now trim the edges of motifs based on a local multiple alignment with gaps. MEME first determines the occurrences of the motif, then aligns each occurrence to the highest-scoring occurrence. These are combined into a multiple alignment, and MEME looks for a set of columns with no gaps. If a set of gapless columns at least <minw> wide is not found, MEME searches for a set with at most 1, 2, ... etc gaps until a set is found.
    • Prior distribution on the number of occurrences: MEME now places a prior on the number of occurrences that controls how strong the bias towards motifs with a given number of sites is. Using a prior improves the performance of MEME with DNA, where the megaprior heuristic (used by default with protein sequences), is not applicable.
  • MAST enhancements
    • Combining DNA strands: MAST now combines the score of a site in a DNA sequence with the score of the corresponding site on the reverse complement strand. (The final score for the site is the maximum of the two scores.)
    • Scoring DNA strands separately: MAST still allows each strand of a DNA sequence to be treated as a separate sequence at the user's request.
    • Ignoring reverse complement DNA: MAST also allows the user to score only the given DNA strand, not scoring the reverse complement strand at all.
    • Background model: MAST now allows the user to specify the background residue frequencies used for computing E-values of scores.
    • Composition-adjusted statistics: MAST can now use a different random model for each target sequence, based on the letter composition of that sequence. This can greatly reduce erroneous matches due to biased sequence composition.
    • New databases: The MAST website includes several new databases including "upstream" sequences for E. coli, B. subtilis and S. cerevesiae, and many complete genomes from GenBank.
  • MEME and MAST enhancements
    • MEME and MAST can be installed under the following operating systems:
      • Mac OS X
      • Linux (various manufacturers)
      • SunSparc workstations (SunOS and Solaris operating systems)
      • Decalpha workstations (OSF/1 operating system)
      • Silicon Graphics workstations (IRIX version 5.3 operating system)
      • Intel Paragon parallel computers
      • Cray T3D parallel computers
      • IBM SP (AIX operating system)

MEME version 2.2 -- February 25, 1998

  • MEME enhancements
    • Sequence weights can be given in the input sequence file.
    • The sites composing each motif are output in BLOCKS or FASTA format. BLOCKS format motifs can be converted to LOGOS, PSSMS and phylogeny trees using the Blocks Multiple Alignment Processor.
    • A "negative" dataset may be specified causing the motifs to be optimized to discriminate between the training set family and the negative family. (This option is only available when you install MEME on your own computer.)
    • An advanced version of the MEME data submission form allows
      • discovering palindromic DNA motifs,
      • discovering motifs occurring on both DNA strands, and
      • discovering more than ten motifs.
  • MAST enhancements
    • DNA sequences (translated in six reading frames) can be searched using protein motifs.
    • Hypertext (HTML) output including links to the ENTREZ database and improved motif diagrams are now output.
    • Additional searchable databases added to the MAST website.
    • The annotation section of output is smaller now since only regions in sequences where the motifs are present are printed.
    • The ambiguous characters "*" and "-" are now permitted in motifs and sequence databases. They are treated as a single, unknown character in databases, and replaced by a weighted average of scores in motifs.

MEME version 2.1 -- March 25, 1997

  • Enhancements
    • MAST output now includes measurements of similarities between all pairs of motifs in the query and a warning if any motifs are too similar. (Too similar motifs in a query can cause some p-values and e-values to be underestimated. These motifs can be removed from the query and MAST re-run to avoid the problem.)
    • Cray T3E parallel computer now supported for MEME and MAST.

MEME version 2.0 -- October 17, 1996

  • Enhancements
    • The output format of MEME has been improved to aid readability and user-friendliness.
    • MEME and MAST have been compiled and tested on:
      • SunSparc workstations (SunOS and Solaris operating systems)
      • Decalpha workstations (OSF/1 operating system)
      • Silicon Graphics workstations (IRIX version 5.3 operating system)
      • Intel Paragon parallel computers
      • Cray T3D parallel computers
    • MAST has been completely rewritten to provide more accurate p-values and to run significantly faster.
    • A MAST server has been added to the MEME system website.
    • The MAST server lets you use the motifs found by MEME in your sequences to search a wide variety of sequence databases including:
      • nr (non-redundant protein and nucleotide databases)
      • month (new or revised sequences in the last 30 days)
      • swissprot (the last major release of the SWISS-PROT protein sequence database)
      • genpept (protein translations from the GenBank feature table)
      • dbest (expressed sequence tag database)
      • dbsts (sequence tag site database)
    • Several improvements have been made to the MEME website:
      • You can now give the name of a file containing your sequences instead of having to cut-and-paste the sequences themselves.
      • You may ask MEME to favor wide motifs instead of the (default) narrow motifs; this is especially useful with small sequence sets (fewer than 10 sequences.)
      • The on-line documentation of MEME has been expanded and improved.
      • MEME now determines the type of your sequences (PROTEIN or DNA) automatically.
      • The MEME server now runs MAST on your sequences to make it easy for you to see the ordering and spacing of the motifs MEME discovered.
    • NOTE and PROBER have been removed from the MEME system since their functionality is now superseded by MAST.

MEME version 1.4 -- February 29, 1996

  • Megaprior heuristic added to MEME.
    • MEME now uses the megaprior heuristic with TCM models and the modified megaprior heuristic with ZOOPS models (by default). This greatly improves the sensitivity and selectivity of motifs using these types of models.
  • P-value computation added to MAST.
    • MAST now computes p-values for sequences when searching large databases. This provides better discrimination between true and false positives than using z-scores. MAST is still only available via ftp, not via the web server.
  • Various bug fixes to MEME.

MEME version 1.3 -- December 18, 1995

  • MEME web server introduced!
  • Various bug fixes to MEME.

MEME version 1.2 -- November 27, 1995

  • MAST homology searcher introduced!
    • MAST allows accurate homology searches of databases using motifs generated by MEME. The scores from all motifs characterizing a family are combined, normalized for sequence length, and sorted by z-score. The output of MAST is similar to that of BLAST, but, used in conjunction with MEME, MAST allows searching for sequences related to an entire family, not just a single sequence in the family. This provides better sensitivity and selectivity than single-sequence homology searches. MAST is currently only available in the ftp-able code, not via the web server.
  • Various bug fixes to MEME.

MEME version 1.1 -- June 30, 1995

  • Initial release of MEME, NOTE and PROBER.

Meta-MEME Release Notes

Release 3.3
  • The program motif-scan has been renamed to fimo

Release 3.x, Release 3.3, March 24, 2007
  • Added new tools:
    Scans a database of sequences for the presence of motifs by calculating the p-value of the match to the motif at each position in the sequences.
    Calculates the log-odds of a multiple alignment for a given phylogentic tree and evolutionary model at each position in the alignment.
    Scans a multiple alignment for the presence of motifs by by calculating the p-values of motif-width windows in the alignment for a given phylogentic tree and evolutionary model.
    Searches target motifs for elements similar to any in a set of query motifs.
  • Added --global option to mhmms. This option causes mhmms to score sequences using the best match between the model and the entire sequence. The default is to score using the best local match within a sequence.
  • Added --maxhits option to mhmms. This option sets an absolute limit to the number of hits returned by mhmms. The default is to return all hits consistent with the E-value and p-value thresholds.
  • Fixed error in mhmms that resulted in a sequence being considered a hit even if all positions were matched to a spacer state.
  • Fixed error causing segmentation a fault in mhmms when the --motif-scoring or --pthresh options were used.
  • Fixed error causing segmentation a fault in mhmm when the --order option was used.

Release 3.21, August 3, 2005
  • Modified meme-io.c to be able to parse the output from the latest versions of MEME.
  • Modified the mcast script: removed the -allow-weak-motifs swich, and added the --keep-unused switch to the call to mhmm from mcast.

Release 3.2, October 7. 2004
  • Reorganized source directory structure and implemented build system using autoconf and automake.
  • Updated program documentation.

Release 3.1, February 4, 2004
  • Remove the auxiliary program score-n-store.
  • Change command-line processing to follow POSIX.2 standard.

Release 3.0.1, June 2, 2003
  • Add documentation for MCAST.

Release 3.0, May 15, 2003
  • Remove mhmmt (EM training program) and mhmma (multiple alignment program).
  • Add mhmmscan and the mcast wrapper program for searching DNA databases for regulatory modules.
  • Allow star topology in mhmm.