This summary table gives the main attributes of the motif.

E-value

The statistical significance of the motif. MEME usually finds the most statistically significant (low E-value) motifs first. The E-value of a motif is based on its log likelihood ratio, width, sites, the background letter frequencies (given in the command line summary), and the size of the training set. The E-value is an estimate of the expected number of motifs with the given log likelihood ratio (or higher), and with the same width and site count, that one would find in a similarly sized set of random sequences. (In random sequences each position is independent with letters chosen according to the background letter frequencies.)

Width

The width of the motif. Each motif describes a pattern of a fixed with as no gaps are allowed in MEME motifs.

Sites

The number of sites contributing to the construction of the motif.

Log Likelihood Ratio

The log likelihood ratio of the motif.The log likelihood ratio is the logarithm of the ratio of the probability of the occurrences of the motif given the motif model (likelihood given the motif) versus their probability given the background model (likelihood given the null model). (Normally the background model is a 0-order Markov model using the background letter frequencies, but higher order Markov models may be specified via the -bfile option to MEME.)

Information Content

The information content of the motif in bits. It is equal to the sum of the uncorrected information content, R(), in the columns of the LOGO. This is equal relative entropy of the motif relative to a uniform background frequency model.

Relative Entropy

The relative entropy of the motif, computed in bits and relative to the background letter frequencies given in the command line summary. It is equal to the log-likelihood ratio (llr) divided by the number of contributing sites of the motif times 1/ln(2),

re = llr / (sites * ln(2)).

MEME motifs are represented by position-specific probability matrices that specify the probability of each possible letter appearing at each possible position in an occurrence of the motif. These are displayed as "sequence LOGOS", containing stacks of letters at each position in the motif. The total height of the stack is the "information content" of that position in the motif in bits. The height of the individual letters in a stack is the probability of the letter at that position multiplied by the total information content of the stack.

Note: The MEME LOGO differs from those produced by the Weblogo program because a small-sample correction is NOT applied. However, MEME LOGOs in PNG and encapsulated postscript (EPS) formats with small-sample correction (SSC) are available by clicking on the download button with "SSC" set to "on" under Download LOGO. The MEME LOGOs without small sample correction are similarly available. Error bars are included in the LOGOs with small-sample correction.

Modern web browsers supporting the canvas element and it's text manipulation functions as described in the html 5 standard, can render the sequence LOGOs without needing the images. The browsers which work with this feature are:

• Firefox 3.5 and above
• Safari 4 and above
• Google Chrome 4 and above

Unfortunately Internet Explorer 8 does not support any html 5 features.

The information content of each motif position is computed as described in the paper by Schneider and Stephens, "Sequence Logos: A New Way to Display Consensus Sequences" but the small-sample correction, e(n), is set to zero for the LOGO displayed in the MEME output. The corrected information content of position i is given by

            R(i) for amino acids   = log2(20) - (H(i) + e(n))   (1a) 
            R(i) for nucleic acids =    2    - (H(i) + e(n))    (1b)
          

where H(i) is the entropy of position i,

            H(l) = - (Sum f(a,i) * log2[ f(a,i) ]).             (2)
          

Here, f(a,i) is the frequency of base or amino acid a at position i, and e(n) is the small-sample correction for an alignment of n letters. The height of letter a in column i is given by

            height = f(a,i) * R(i)                              (3)
          

The approximation for the small-sample correction, e(n), is given by:

            e(n) = (s-1) / (2 * ln(2) * n),                     (4)
          

where s is 4 for nucleotides, 20 for amino acids, and n is the number of sequences in the alignment.

The letters in the logos are colored as follows.
For DNA sequences, the letter categories contain one letter each.

For proteins, the categories are based on the biochemical properties of the various amino acids.

J. Kyte and R. Doolittle, 1982. "A Simple Method for Displaying the Hydropathic Character of a Protein", J. Mol Biol. 157, 105-132.

Note: the "text" output format of MEME preserves the historical MEME format where LOGOS are replaced by a simplified probability matrix, a relative entropy plot, and a multi-level consensus sequence.

Logos can be generated on the fly by the meme webservice and you may specify a number of options to customize them to your needs. The options are:

Orientation

Only valid for nucleotide motifs. Generate the standard view or the reverse complemented view of the motif.

SSC

Use small sample correction and show errorbars on the image. Small sample correction is used by the Weblogo program.

Format

The format of the generated image. If the image is to be used on the web then png is recommend. If the image is to be published then eps is recommended.

Width

The width of the generated image in centimetres.

Height

The height of the generated image in centimetres.

This is a regular expression (RE) describing the motif. In each column, all letters with observed frequencies greater than 0.2 are shown; less-frequent letters are not included in the RE. MEME regular expressions are interpreted as follows: single letters match that letter; groups of letters in square brackets match any of the letters in the group. Regular expressions can be used for searching for the motif in sequences (using, for example, PatMatch) but the search accuracy will usually be better with the PSSM (using, for example MAST.)

Either as a group or individually the motifs have a number of options for further analysis.

MAST

Finds the best matching positions for a set of motifs in each sequence provided to it, ranked by the combined score of each sequence. For more information about MAST please read the introduction.

FIMO

Finds all matches for a motif. For more information about FIMO please read the introduction.

TOMTOM

Compares a single motif to a database of motifs. For more information about TOMTOM please read the introduction.

GOMO

Identifies possible roles of DNA binding motifs. For more information about GOMO please read the introduction.

BLOCKS

Submit to Blocks Multiple Alignment Processor where you can do several things like create phylogeny trees and search the blocks against a database of other blocks (protein only). For more information about BLOCKS Multiple Alignment Processor please visit the website.

The extracted data is avaliable in the following formats.

PSPM Format

The motif itself is a position-specific probability matrix giving, for each position in the pattern, the observed frequency ("probability") of each possible letter. The probability matrix is printed "sideways"--columns correspond to the letters in the alphabet (in the same order as shown in the simplified motif) and rows corresponding to the positions of the motif, position one first. The motif is preceded by a line starting with "letter-probability matrix:" and containing the length of the alphabet, width of the motif, number of occurrences of the motif, and the E-value of the motif.
Note: Earlier versions of MEME gave the posterior probabilities--the probability after applying a prior on letter frequencies--rather than the observed frequencies. These versions of MEME also gave the number of possible positions for the motif rather than the actual number of occurrences. The output from these earlier versions of MEME can be distinguished by "n=" rather than "nsites=" in the line preceding the matrix.

PSSM Format

The position-specific scoring matrix corresponding to the motif is printed for use by database search programs such as MAST. This matrix is a log-odds matrix calculated by taking 100 times the log (base 2) of the ratio p/f at each position in the motif where p is the probability of a particular letter at that position in the motif, and f is the background frequency of the letter (given in the command line summary section.) This is the same matrix that is used above in computing the p-values of the occurrences of the motif in the Sites and Block Diagrams sections. The scoring matrix is printed "sideways"--columns correspond to the letters in the alphabet (in the same order as shown in the simplified motif) and rows corresponding to the positions of the motif, position one first. The scoring matrix is preceded by a line starting with "log-odds matrix:" and containing the length of the alphabet, width of the motif, number of characters in the training set, the scoring threshold (obsolete) and the motif E-value.
Note: The probability p used to compute the PSSM is not exactly the same as the corresponding value in the Position Specific Probability Matrix (PSPM). The values of p used to compute the PSSM take into account the motif prior, whereas the values in the PSPM are just the observed frequencies of letters in the motif sites.

BLOCKS Format

For use with BLOCKS tools.

FASTA Format

The FASTA format as described here.

Raw Format

Just the sites of the sequences that contributed to the motif. One site per line.

MEME displays the occurrences (sites) of the motif in the training set. The sites are shown aligned with each other, and the ten sequence positions preceding and following each site are also shown. Each site is identified by the name of the sequence where it occurs, the strand (if both strands of DNA sequences are being used), and the position in the sequence where the site begins. When the DNA strand is specified, '+' means the sequence in the training set, and '-' means the reverse complement of the training set sequence. (For '-' strands, the 'start' position is actually the position on the positive strand where the site ends.) The sites are listed in order of increasing statistical significance (p-value). The p-value of a site is computed from the the match score of the site with the position specific scoring matrix for the motif. The p-value gives the probability of a random string (generated from the background letter frequencies) having the same match score or higher. (This is referred to as the position p-value by the MAST algorithm.)

The occurrences of the motif in the training set sequences are shown as coloured blocks on a line. One diagram is printed for each sequence showing all the sites contributating to that motif in that sequence. The sequences are listed in the same order as in the input to make it easier to compare multiple block diagrams. Additionally the best p-value for the sequence/motif combination is listed though this may not be in ascending order as with the sites. The p-value of an occurrence is the probability of a single random subsequence the length of the motif, generated according to the 0-order background model, having a score at least as high as the score of the occurrence. When the DNA strand is specified '+', it means the motif appears from left to right on the sequence, and '-' means the motif appears from right to left on the complementary strand. A sequence position scale is shown at the end of each table of block diagrams.

The motif occurrences shown in the motif summary may not be exactly the same as those reported in each motif section because only motifs with a position p-value of 0.0001 that don't overlap other, more significant motif occurrences are shown.

See the documentation for MAST output for the definition of position and combined p-values.