| 1 |
# |
| 2 |
# parameters for all Drosophila versions |
| 3 |
# |
| 4 |
# date : 11.8.2009 |
| 5 |
# |
| 6 |
|
| 7 |
# |
| 8 |
# Properties for augustus |
| 9 |
#------------------------------------ |
| 10 |
/augustus/verbosity 3 # 0-3, 0: only print the necessary |
| 11 |
maxDNAPieceSize 200000 # maximum segment that is predicted in one piece |
| 12 |
stopCodonExcludedFromCDS false # make this 'true' if the CDS includes the stop codon (training and prediction) |
| 13 |
|
| 14 |
# gff output options: |
| 15 |
protein on # output predicted protein sequence |
| 16 |
codingseq off # output the coding sequence |
| 17 |
cds on # output 'cds' as feature for exons |
| 18 |
start on # output start codons (translation start) |
| 19 |
stop on # output stop codons (translation stop) |
| 20 |
introns on # output introns |
| 21 |
tss on # output transcription start site |
| 22 |
tts on # output transcription termination site |
| 23 |
print_utr off # output 5'UTR and 3'UTR lines in addition to exon lines |
| 24 |
|
| 25 |
checkExAcc off # internal parameter for extrinsic accuracy |
| 26 |
|
| 27 |
# alternative transcripts and posterior probabilities |
| 28 |
sample 100 # the number of sampling iterations |
| 29 |
alternatives-from-sampling false # output alternative transcripts |
| 30 |
minexonintronprob 0.08 # minimal posterior probability of all (coding) exons |
| 31 |
minmeanexonintronprob 0.4 # minimal geometric mean of the posterior probs of introns and exons |
| 32 |
maxtracks -1 # maximum number of reported transcripts per gene (-1: no limit) |
| 33 |
keep_viterbi true # set to true if all Viterbi transcripts should be reported |
| 34 |
uniqueCDS true # don't report transcripts that differ only in the UTR |
| 35 |
UTR off # predict untranslated regions |
| 36 |
|
| 37 |
# |
| 38 |
# |
| 39 |
# The rest of the file contains mainly meta parameters used for training. |
| 40 |
# |
| 41 |
|
| 42 |
# global constants |
| 43 |
# ---------------------------- |
| 44 |
|
| 45 |
/Constant/trans_init_window 25 |
| 46 |
/Constant/ass_upwindow_size 32 |
| 47 |
/Constant/ass_start 1 |
| 48 |
/Constant/ass_end 4 |
| 49 |
/Constant/dss_start 3 |
| 50 |
/Constant/dss_end 4 |
| 51 |
/Constant/init_coding_len 9 |
| 52 |
/Constant/intterm_coding_len 0 |
| 53 |
/Constant/tss_upwindow_size 45 |
| 54 |
/Constant/decomp_num_at 1 |
| 55 |
/Constant/decomp_num_gc 1 |
| 56 |
/Constant/gc_range_min 0.32 # This range has an effect only when decomp_num_steps>1. |
| 57 |
/Constant/gc_range_max 0.50 # States the minimal and maximal percentage of c or g |
| 58 |
/Constant/decomp_num_steps 1 # I recommend keeping this to 1 for most species. |
| 59 |
/Constant/min_coding_len 201 # no gene with a coding sequence shorter than this is predicted |
| 60 |
/Constant/probNinCoding 0.23 # divide this by .25 to get a malus for making one masked letter part of the coding sequence |
| 61 |
/Constant/amberprob 0.34 # Prob(stop codon = tag), if 0 tag is assumed to code for amino acid |
| 62 |
/Constant/ochreprob 0.41 # Prob(stop codon = taa), if 0 taa is assumed to code for amino acid |
| 63 |
/Constant/opalprob 0.25 # Prob(stop codon = tga), if 0 tga is assumed to code for amino acid |
| 64 |
/Constant/subopt_transcript_threshold 0.7 |
| 65 |
/Constant/almost_identical_maxdiff 10 |
| 66 |
|
| 67 |
# type of weighing, one of 1 = equalWeights, 2 = gcContentClasses, 3 = multiNormalKernel |
| 68 |
/BaseCount/weighingType 3 |
| 69 |
# file with the weight matrix (only for multiNormalKernel type weighing) |
| 70 |
/BaseCount/weightMatrixFile fly_exp_weightmatrix.txt # change this to your species if at all necessary |
| 71 |
|
| 72 |
# Properties for IGenicModel |
| 73 |
# ---------------------------- |
| 74 |
/IGenicModel/verbosity 0 |
| 75 |
/IGenicModel/infile fly_exp_igenic_probs.pbl # change this and the other five filenames *_probs.pbl below to your species |
| 76 |
/IGenicModel/outfile fly_exp_igenic_probs.pbl |
| 77 |
/IGenicModel/patpseudocount 5.0 |
| 78 |
/IGenicModel/k 4 # order of the Markov chain for content model, keep equal to /ExonModel/k |
| 79 |
|
| 80 |
# Properties for ExonModel |
| 81 |
# ---------------------------- |
| 82 |
/ExonModel/verbosity 3 |
| 83 |
/ExonModel/infile fly_exp_exon_probs.pbl |
| 84 |
/ExonModel/outfile fly_exp_exon_probs.pbl |
| 85 |
/ExonModel/patpseudocount 5.0 |
| 86 |
/ExonModel/minPatSum 350 |
| 87 |
/ExonModel/k 4 # order of the Markov chain for content model |
| 88 |
/ExonModel/etorder 2 |
| 89 |
/ExonModel/etpseudocount 3 |
| 90 |
/ExonModel/exonlengthD 3000 # beyond this the distribution is geometric |
| 91 |
/ExonModel/maxexonlength 15000 |
| 92 |
/ExonModel/slope_of_bandwidth 0.3 |
| 93 |
/ExonModel/minwindowcount 8 |
| 94 |
/ExonModel/tis_motif_memory 3 |
| 95 |
/ExonModel/tis_motif_radius 2 |
| 96 |
|
| 97 |
# Properties for IntronModel |
| 98 |
# ---------------------------- |
| 99 |
/IntronModel/verbosity 0 |
| 100 |
/IntronModel/infile fly_exp_intron_probs.pbl |
| 101 |
/IntronModel/outfile fly_exp_intron_probs.pbl |
| 102 |
/IntronModel/patpseudocount 5.0 |
| 103 |
/IntronModel/k 4 # order of the Markov chain for content model, keep equal to /ExonModel/k |
| 104 |
/IntronModel/slope_of_bandwidth 0.4 |
| 105 |
/IntronModel/minwindowcount 3 |
| 106 |
/IntronModel/asspseudocount 0.01 |
| 107 |
/IntronModel/dsspseudocount 0.01015 |
| 108 |
/IntronModel/dssneighborfactor 0.001 |
| 109 |
#/IntronModel/splicefile fly_splicefile.txt # this optional file contains additional windows around splice sites for training, uncomment if you have one |
| 110 |
/IntronModel/sf_with_motif false # if true the splice file is also used to train the branch point region |
| 111 |
/IntronModel/d 929 # constraint: this must be larger than 4 + /Constant/dss_end + /Constant/ass_upwindow_size + /Constant/ass_start |
| 112 |
/IntronModel/ass_motif_memory 1 |
| 113 |
/IntronModel/ass_motif_radius 4 |
| 114 |
|
| 115 |
# Properties for UtrModel |
| 116 |
# ---------------------------- |
| 117 |
/UtrModel/verbosity 3 |
| 118 |
/UtrModel/infile fly_exp_utr_probs.pbl |
| 119 |
/UtrModel/outfile fly_exp_utr_probs.pbl |
| 120 |
/UtrModel/k 4 |
| 121 |
/UtrModel/utr5patternweight 0.3 #0.7625 |
| 122 |
/UtrModel/utr3patternweight 0.3 #0.5 |
| 123 |
/UtrModel/patpseudocount 1 |
| 124 |
/UtrModel/tssup_k 1 |
| 125 |
/UtrModel/tssup_patpseudocount 1 |
| 126 |
/UtrModel/slope_of_bandwidth 0.25 |
| 127 |
/UtrModel/minwindowcount 1 |
| 128 |
/UtrModel/exonlengthD 800 |
| 129 |
/UtrModel/maxexonlength 1200 |
| 130 |
/UtrModel/max3singlelength 2000 # excludes roughly 1% |
| 131 |
/UtrModel/max3termlength 1200 # excludes ~ 0.3% |
| 132 |
/UtrModel/tss_start 8 |
| 133 |
/UtrModel/tss_end 5 |
| 134 |
/UtrModel/tata_start 2 |
| 135 |
/UtrModel/tata_end 10 |
| 136 |
/UtrModel/tata_pseudocount 2 |
| 137 |
/UtrModel/d_tss_tata_min 26 # minimal distance between start of tata box (if existent) and tss |
| 138 |
/UtrModel/d_tss_tata_max 37 # maximal distance between start of tata box (if existent) and tss |
| 139 |
/UtrModel/polyasig_consensus aataaa # polyadenylation signal training not fully automated yet |
| 140 |
/UtrModel/d_polyasig_cleavage 14 # the transcription end is predicted this many bases after the polyadenylation signal |
| 141 |
/UtrModel/d_polya_cleavage_min 9 |
| 142 |
/UtrModel/d_polya_cleavage_max 35 |
| 143 |
/UtrModel/prob_polya 0.95 |
| 144 |
/UtrModel/tts_motif_memory 1 |