wget -c http://homer.ucsd.edu/homer/configureHomer.pl
homer --help
homer : Empirical Motif Optimizer
usage: ./homer [data] [parameters] -a [action]
This program is meant to be called from other programs (i.e. findMotifsGenome.pl), and not used directly
Data options:
-dna|-prot : Sequence type (-dna)
-s <file> : Sequence File
-g <file> : Group/Stat File
-mer <file> : Mer File
-m <file> : PSSM Motif File
-o <file> : output file prefix
-seed <file> : seed file
-offset <#> : offset of sequence from TSS (-2000)
Parameter options:
-exact : remember mapping between mers and genes (default: approx)
-w : Weight sequences (according to addition columns in group file: 1st-gene 2nd-sequence)
-T : Test all sequences as candidate motifs (default: only test target sequences)
-noautoscale : Do not autoscale sequences to be equal in foreground and background
-freqAdjust : Compute log-odds using frequency, default (0.25)
-dual : find dual motifs in the form A<gap>B where A and B can be rev-opposites
-flip : find dual motifs in the form A<gap>B or B<gap>A
-zoopsapprox <OFF,#(max to count)> : (counts multiple motifs per sequence | default: 2)
-norevopp : don't search opposite strand (default->DNA:yes, Protein:no)
-min <#> : min mer size (6)
-max <#> : max mer size [also standard mer size] (10)
-len <#> : Find motifs of length # (default=10)
-gap <#,#,#-#> : Find motifs with gaps(0)(i.e. -gap 3 -gap 2,4,5 -gap 1-10
Gaps will only be in the center of motif and will only use even lengthed motifs
-mis <#> : # of mismatches to check for degeneracy (1)
-IUPAC <#> : # of IUPAC codes per mer that can be used during global optimization (0)
-iupactype <1,2,or3> : Type of IUPAC symbols used
1: (default) Only N is used
2: Only N and 2 bp symbols are used (i.e. R = A or G
3: Full IUPAC code is used (includes 3-way symbols)
-S <#> : number of seeds to check during profile optimization(50)
-branch <#> : sets depth of optimization (closer to zero the more sensitive (0.5))
-I <#> : maximum number of iterations during optimization (5)
-rmalign : DO NOT remove aligned seeds
-maxneg <0 to 1> maximum percentage of negative genes that can contain the motif
-speed <NORMAL|FAST>: Program will heuristically avoid performing exhaustive
calculations (default: FAST)
Scoring Functions:
-alg <method> : scoring algorithm (default: hypergeo)
hypergeo - hypergeometric scoring (ZOOPS)
binomial - binomical scoring [for variable length seq] (ZOOPS) (requires exact)
approxbinomial - binomical scoring [for variable length seq] (ZOOPS) (requires exact)
sitehypergeo - hypergeometric scoring across seq positions (very slow)
sitebinomial - binomial scoring across seq positions
fisher <#> - fisher exact test (slow, # scales exponentially)
<# = largest repetition to consider [default=2]>
rank - group file must have sortable numeric value
freqdiff - used by most bayesian/nnet programs
logit - used by most bayesian/nnet programs
Background Modeling options (this forces a binomial style scoring function):
-b <method> [method options...]
markov <#> - generate hmm from target sequences using a hmm of order #
bmarkov <#> - generate hmm from background sequences using a hmm of order #
mosaic - generate mosaic hmm from background sequences **coming soon**
Filter Options:
-N <float> : filtering cutoff for ratio of N's in sequence (0.9)
-seqless <#> : filter sequences shorter than #
-seqmore <#> : filter sequences longer than #
Actions (-a):
MOTIFS - Find motifs <outfile>.motifs# where # = motif length
MERS - Create mer file (low memory) <stdout>
DMERS - Create degenerate mer file <stdout>
FIND - find motifs in sequence <stdout>
OPTPVALUE - optimize motif threshold and pvalue (exact)<stdout>
GETPVALUE - get the p-value enrichment for a given motif(exact)<stdout>
GENESCORE - returns highest motif score for each gene <stdout>
REFINE - optimize motif PSSM profile, threshold, and pvalue <stdout>
REFINETHRESH - optimize motif PSSM threshold and pvalue <stdout>
CLUSTER - cluster mers from seed file (can't use exact scoring) <outfile>
SORTMERS - sort a mer file according to pvalue <stdout>
REMOVE - removes motif from sequence (replaces with N's) <stdout>
This program is meant to be called from other programs (i.e. findMotifsGenome.pl), and not used directly
findMotifsGenome.pl -h
Program will find de novo and known motifs in regions in the genome
Usage: findMotifsGenome.pl <pos file> <genome> <output directory> [additional options]
Example: findMotifsGenome.pl peaks.txt mm8r peakAnalysis -size 200 -len 8
Possible Genomes:
-- or --
Custom: provide the path to genome FASTA files (directory or single file)
Heads up: will create the directory "preparsed/" in same location.
Basic options:
-mask (mask repeats/lower case sequence, can also add 'r' to genome, i.e. mm9r)
-bg <background position file> (genomic positions to be used as background, default=automatic)
removes background positions overlapping with target positions unless -keepOverlappingBg is used
-chopify (chop up large background regions to the avg size of target regions)
-len <#>[,<#>,<#>...] (motif length, default=8,10,12) [NOTE: values greater 12 may cause the program
to run out of memory - in these cases decrease the number of sequences analyzed (-N),
or try analyzing shorter sequence regions (i.e. -size 100)]
-size <#> (fragment size to use for motif finding, default=200)
-size <#,#> (i.e. -size -100,50 will get sequences from -100 to +50 relative from center)
-size given (uses the exact regions you give it)
-S <#> (Number of motifs to optimize, default: 25)
-mis <#> (global optimization: searches for strings with # mismatches, default: 2)
-norevopp (don't search reverse strand for motifs)
-nomotif (don't search for de novo motif enrichment)
-rna (output RNA motif logos and compare to RNA motif database, automatically sets -norevopp)
Scanning sequence for motifs
-find <motif file> (This will cause the program to only scan for motifs)
Known Motif Options/Visualization
-mset <vertebrates|insects|worms|plants|yeast|all> (check against motif collects, default: auto)
-basic (just visualize de novo motifs, don't check similarity with known motifs)
-bits (scale sequence logos by information content, default: doesn't scale)
-nocheck (don't search for de novo vs. known motif similarity)
-mcheck <motif file> (known motifs to check against de novo motifs,
-float (allow adjustment of the degeneracy threshold for known motifs to improve p-value[dangerous])
-noknown (don't search for known motif enrichment, default: -known)
-mknown <motif file> (known motifs to check for enrichment,
-nofacts (omit humor)
-seqlogo (use weblogo/seqlogo/ghostscript to generate logos, default uses SVG now)
Sequence normalization options:
-gc (use GC% for sequence content normalization, now the default)
-cpg (use CpG% instead of GC% for sequence content normalization)
-noweight (no CG correction)
Also -nlen <#>, -olen <#>, see homer2 section below.
Advanced options:
-h (use hypergeometric for p-values, binomial is default)
-N <#> (Number of sequences to use for motif finding, default=max(50k, 2x input)
-local <#> (use local background, # of equal size regions around peaks to use i.e. 2)
-redundant <#> (Remove redundant sequences matching greater than # percent, i.e. -redundant 0.5)
-maxN <#> (maximum percentage of N's in sequence to consider for motif finding, default: 0.7)
-maskMotif <motif file1> [motif file 2]... (motifs to mask before motif finding)
-opt <motif file1> [motif file 2]... (motifs to optimize or change length of)
-rand (randomize target and background sequences labels)
-ref <peak file> (use file for target and background - first argument is list of peak ids for targets)
-oligo (perform analysis of individual oligo enrichment)
-dumpFasta (Dump fasta files for target and background sequences for use with other programs)
-preparse (force new background files to be created)
-preparsedDir <directory> (location to search for preparsed file and/or place new files)
-keepFiles (keep temporary files)
-fdr <#> (Calculate empirical FDR for de novo discovery #=number of randomizations)
homer2 specific options:
-homer2 (use homer2 instead of original homer, default)
-nlen <#> (length of lower-order oligos to normalize in background, default: -nlen 3)
-nmax <#> (Max normalization iterations, default: 160)
-neutral (weight sequences to neutral frequencies, i.e. 25%, 6.25%, etc.)
-olen <#> (lower-order oligo normalization for oligo table, use if -nlen isn't working well)
-p <#> (Number of processors to use, default: 1)
-e <#> (Maximum expected motif instance per bp in random sequence, default: 0.01)
-cache <#> (size in MB for statistics cache, default: 500)
-quickMask (skip full masking after finding motifs, similar to original homer)
-minlp <#> (stop looking for motifs when seed logp score gets above #, default: -10)
Original homer specific options:
-homer1 (to force the use of the original homer)
-depth [low|med|high|allnight] (time spent on local optimization default: med)
参考资料：
http://homer.ucsd.edu/homer/introduction/install.html
[软件使用 2] HOMER安装和使用攻略，如何获取Motif? - 知乎 (zhihu.com)
http://www.360doc.com/content/21/0714/12/76149697_986500345.shtml
网友评论