Click on the menu at the left to see which of the following motif input methods are available.
You may input both probability and count matrices of either orientation and the rules described below will be used to convert the matrix into a MEME formatted motif.
The counts/probabilities are expected to be ordered based on the alphabetical ordering of their codes. So DNA is ordered ACGT and protein is ordered ACDEFGHIKLMNPQRSTVWY. For custom alphabets the ordering goes uppercase letters (A-Z), lowercase letters (a-z), numbers (0-9) and finally the symbols '*', '-' and '.'.
Matrix motifs may be input with either one position per row (preferred) which is called row orientation, or one position per column which is called column orientation. The orientation is determined by picking which dimension (row or column) is equal to the alphabet size. If both dimensions are equal to the alphabet size then row orientation is assumed. If neither dimension is equal to the alphabet size then the closest that is still smaller than the alphabet size is picked, however if both are equally smaller then column orientation is assumed. Finally if none of the above rules work to determine the orientation then row orientation is assumed.
Once the orientation is determined, the sum of the numbers that make up the first position is calculated and rounded to the nearest integer. If that value is larger than 1 then the matrix is assumed to be a count matrix and that value is used as the site count, otherwise the matrix is assumed to be a probability matrix and a site count of 20 is used.
Once the orientation is determined then each number in the matrix is converted to a normalized probability by dividing by the sum of all the numbers for that motif position. If any numbers are missing they are assumed to have the value zero. As a special case if all numbers in a motif position have the value zero then they are given the uniform probability of 1 / alphabet size.
Red asterisks (*) indicate where the parser thinks values are missing. A yellow highlighted row or column with a red number at the end indicates that the counts for that position don't sum to the same count as the first position. The red number shows the difference. If the red number is negative then that position sums to less then the first position, if it is positive then it sums to more than the first position.
You may input one or more sites of the motif including using ambiguity codes or bracket expressions to represent multiple possibilities for a single motif position.
The DNA and protein alphabets include additional codes that represent multiple possible bases. For example the DNA alphabet includes W (for weak) which represents that the given position could be either a A (for adenosine) or a T (for thymidine).
Bracket expressions also group together multiple codes so they share a single position. Their syntax is a opening square bracket '[' followed by one or more codes and a closing square bracket ']'. For example with a DNA motif the bracket expression [AT] means that both A and T are acceptable and is equivalent to the ambiguity code W. Any repeats of a base in a bracket expression are ignored so for example a DNA bracket expression [AAT] has the same effect as [AT] or [AW] or W.
When only one site is provided the site count is set to 20, however you can precisely control the motif by providing multiple sites. Each of these sites can still contain ambiguity codes and bracket expressions but a single count will be divided among the selected bases for each position. When multiple sites are provided the site count will be set to the number of sites provided.
DNA motifs support the standard 4 codes for the bases: adenosine (A), cytidine (C), guanosine (G) and thymidine (T) as well as supporting the following ambiguity codes.
|Not A||B||C, G, T|
|Not C||D||A, G, T|
|Not G||H||A, C, T|
|Not T||V||A, C, G|
|Any||N||A, C, G, T|
Protein motifs support the standard 20 codes for the amino acids: Alanine (A), Arginine (R), Asparagine (N), Aspartic acid (D), Cysteine (C), Glutamic acid (E), Glutamine (Q), Glycine (G), Histidine (H), Isoleucine (I), Leucine (L), Lysine (K), Methionine (M), Phenylalanine (F), Proline (P), Serine (S), Threonine (T), Tryptophan (W), Tyrosine (Y) and Valine (V) as well as supporting the following ambiguity codes.
|Asparagine or aspartic acid||B||N, D|
|Glutamine or glutamic acid||Z||E, Q|
|Leucine or Isoleucine||J||I, L|
|Unspecified or unknown amino acid||X||A, C, D, E, F, G, H, I, K, L, M, N, P, Q, R, S, T, V, W, Y|
Note that the two amino acids Selenocysteine (U) and Pyrrolysine (O) are not supported by the MEME Suite.
Single site motif using ambiguity codes N and R or bracket expressions. These give an approximation of the other motifs below.
Multiple site motif. This lists all 28 sites and gives the same result as the count matrix below.
Count matrix motif showing row and column orientations.
Note that all of these can be used with an identifier and alternate name like these 3 count matrix motifs from Jaspar.
When enabled this field supports selecting motifs from the file with a space separated list of motif identifiers and/or their positions in the file.
Any numbers in the range 1 to 999 are assumed to refer to the position of the selected motif in the file, so the entry "3" always refers to the third motif. Any other entry is assumed to be a motif identifier.
Motif identifiers can not start with a dash and can only contain alphanumeric characters as well as colon ':', underscore '_', dot '.' and dash '-'.
Select the desired motif database.
Consult the motif database documentation for descriptions of all the DNA and RNA motif databases present on this MEME Suite server.
This option can help change the alphabet of motifs from a base alphabet to a derived alphabet.
This might be useful if you need to compare an extended DNA motif with a library of DNA motifs, or if you wish to compare RNA motifs to DNA motifs. Note that this option will also let you do nonsensical things like compare Protein motifs to DNA motifs so use it with care.
The derived alphabet must have all the core symbols of the alphabet that it is derived from. For example if the alphabet is derived from DNA it must have ACGT as core symbols. Expanding the alphabet adds frequencies of zero for every symbol in the derived alphabet that did not exist in the base alphabet.
Select an available sequence database from this menu.
Select an available version of the sequence database from this menu.
Select an available tissue/cell-specificity from this menu.
Selecting this option will filter the sequence menu to only contain databases that have additional information that is specific to a tissue or cell line.
This option causes MEME Suite to use tissue/cell-specific information (typically from DNase I or histone modification ChIP-seq data) encoded as a position specific prior that has been created by the MEME Suite create-priors utility. You can see a description of the sequence databases for which we provide tissue/cell-specific priors here.
Note that you cannot upload or type in your own sequences when tissue/cell-specific scanning is selected.
Enter text naming or describing this analysis. The job description will be included in the notification email you receive and in the job output.
You provide one set of sequences MEME-ChIP reports motifs enriched in this set. MEME discovers motifs enriched relative to a random model based on frequencies of the letters in your sequences, or relative to the frequencies given in a "background model" that you may provide (see "Universal options"). DREME discovers motifs that are enriched in your sequences relative to a control set of sequences that DREME creates by shuffling each of your sequences, conserving dinucleotide frequencies. CentriMo identifies motifs enriched in a central (or uncentered, see "CentriMo options") region relative to the flanking regions.
You provide two sets of sequences and MEME-ChIP reports motifs that are enriched in the first (primary) set relative to the second (control) set. Note:The sequences in the primary and control sets should all be the same length.
The primary sequences should all be the same length.
The recommended length for ChIP-seq sequences is 500 bp
centered on the summit (or center if the summit is not known) of a peak.
MEME-ChIP works best with sequences no longer than 2000 bp.
MEME-ChIP can analyze peak regions identified by ChIP-seq, cross-linking sites identified by CLIP-seq and related assays, as well as sets of genomic regions selected using other criteria (e.g., TSSs of differentially expressed or bound genes).
The control sequences should all be the same length as the primary sequences. The recommended length for ChIP-seq sequences is 500 bp centered on the summit (or center if the summit is not known) of a peak. MEME-ChIP works best with sequences no longer than 2000 bp. If the primary sequences are ChIP-seq peak regions from a transcription factor ChIP-seq experiment, similar regions from a knockout cell line or organism, are a possible choice for control sequences. The control sequences should be prepared in exactly the same way (e.g., repeat-masking) as the primary sequences.
MEME-ChIP will use this set of motifs for motif enrichment analysis and will also report if any motifs that it discovers in your sequences are similar to any motifs in this set.
When using the RNA sequences you should select this option to ensure that only the given strand is scanned for motifs.
If your input sequences are sorted in order of confidence (best to worst) then you should select this option. MEME-ChIP will pass the sequences at the start of the file to MEME. By default, MEME-ChIP passes a random subset of your sequences to MEME.
The background model normalizes for biased distribution of letters and groups of letters in your sequences. A 0-order model adjusts for single letter biases, a 1-order model adjusts for dimer biases (e.g., GC content in DNA sequences), etc.
By default MEME-ChIP will determine the background Markov model from the input sequences. You may select a Markov model order of 0 to 4. Alternately you may select "Upload background model" and input a file containing a background model.
The downloadable version of the MEME Suite contains a program named "fasta-get-markov" that you can use to create background model files in the correct format from FASTA sequence files.
This is where you tell MEME how you believe occurrences of the motifs are distributed among the sequences. Selecting the correct type of distribution improves the sensitivity and quality of the motif search.
MEME will keep searching until it finds this many motifs or it hits some other threshold like the maximum run time. Note that MEME does not use an E-value threhold like DREME so you should always check the E-value of any found motifs.
This is the width (number of characters in the sequence pattern) of a single motif. MEME chooses the optimal width of each motif individually using a statistical heuristic function. You can choose different limits for the minimum and maximum motif widths that MEME will consider. The width of each motif that MEME reports will lie within the limits you choose.
This is the total number of sites in the training set where a single motif occurs. You can choose different limits for the minimum and maximum number of occurrences that MEME will consider. If you have prior knowledge about the number of occurrences that motifs have in your training set, limiting MEME's search in this way can can increase the likelihood of MEME finding true motifs.
MEME chooses the number of occurrences to report for each motif by optimizing a statistical heuristic function, restricting the number of occurrences to the range you give here, or using defaults described below if you leave these fields deselected.
|Zero or one occurrence per sequence||sqrt(n)||n|
|One occurrence per sequence||n||n|
|Any number of repetitions||sqrt(n)||min(5*n, 600)|
Checking this box causes MEME to search only for DNA palindromes. This causes MEME to average the letter frequencies in corresponding motif columns together. For instance, if the width of the motif is 10, columns 1 and 10, 2 and 9, 3 and 8, etc., are averaged together. The averaging combines the frequency of A in one column with T in the other, and the frequency of C in one column with G in the other. If this box is not checked, the columns are not averaged together.
DREME stops looking for motifs when one of these limits is met. There is an additional time limit which is set by the server operator.
Specify a minimum score for a match to be considered. If a sequence does not have any matches which meet this minimum score for a given motif, then that sequence will not be considered for that motif.
This option limits the maximum region size that CentriMo will test. This option is useful if your sequences are quite long (> 500 bp) or you are interested only in narrow regions of enrichment. Limiting the size of the maximum region reduces the impact of the multiple testing correction, increasing the sensitivity of the analysis. When this option is not supplied CentriMo will test region sizes up to one less than the maximum number of places that a given motif can align to the sequence.
This is the E-value threshold CentriMo uses for reporting enriched central regions for motifs. If multiple enriched regions overlap then the region with the best p-value and smallest size will be output.
This option causes all regions up to the maximum region size to be considered even if they are not in the center. This can be useful when your sequences are aligned on a genomic landmark (e.g., TSS) since a motif might be enriched at a particular distance upstream or downstream of the landmark.
This option causes CentriMo to store the identifiers (IDs) of sequences that have their best match in the most enriched region for each motif. This will allow you to easily extract the IDs of the sequences contributing to the enrichment of one or more motifs in the CentriMo output. This option makes the CentriMo output file much larger.