mspire 0.2.4 → 0.3.0
Sign up to get free protection for your applications and to get access to all the features.
- data/INSTALL +1 -0
- data/README +25 -0
- data/Rakefile +129 -40
- data/bin/{find_aa_freq.rb → aafreqs.rb} +2 -2
- data/bin/bioworks_to_pepxml.rb +1 -0
- data/bin/fasta_shaker.rb +1 -96
- data/bin/filter_and_validate.rb +5 -0
- data/bin/{mzxml_to_lmat.rb → ms_to_lmat.rb} +8 -7
- data/bin/prob_validate.rb +6 -0
- data/bin/raw_to_mzXML.rb +2 -2
- data/bin/srf_group.rb +1 -0
- data/bin/srf_to_sqt.rb +40 -0
- data/changelog.txt +68 -0
- data/lib/align/chams.rb +6 -6
- data/lib/align.rb +4 -3
- data/lib/bsearch.rb +120 -0
- data/lib/fasta.rb +318 -86
- data/lib/group_by.rb +10 -0
- data/lib/index_by.rb +11 -0
- data/lib/merge_deep.rb +21 -0
- data/lib/{spec → ms/converter}/mzxml.rb +77 -109
- data/lib/ms/gradient_program.rb +171 -0
- data/lib/ms/msrun.rb +209 -0
- data/lib/{spec/msrun.rb → ms/msrun_index.rb} +7 -40
- data/lib/ms/parser/mzdata/axml.rb +12 -0
- data/lib/ms/parser/mzdata/dom.rb +160 -0
- data/lib/ms/parser/mzdata/libxml.rb +7 -0
- data/lib/ms/parser/mzdata.rb +25 -0
- data/lib/ms/parser/mzxml/axml.rb +11 -0
- data/lib/ms/parser/mzxml/dom.rb +159 -0
- data/lib/ms/parser/mzxml/hpricot.rb +253 -0
- data/lib/ms/parser/mzxml/libxml.rb +15 -0
- data/lib/ms/parser/mzxml/regexp.rb +122 -0
- data/lib/ms/parser/mzxml/rexml.rb +72 -0
- data/lib/ms/parser/mzxml/xmlparser.rb +248 -0
- data/lib/ms/parser/mzxml.rb +175 -0
- data/lib/ms/parser.rb +108 -0
- data/lib/ms/precursor.rb +10 -0
- data/lib/ms/scan.rb +81 -0
- data/lib/ms/spectrum.rb +193 -0
- data/lib/ms.rb +10 -0
- data/lib/mspire.rb +4 -0
- data/lib/roc.rb +61 -1
- data/lib/sample_enzyme.rb +31 -8
- data/lib/scan_i.rb +21 -0
- data/lib/spec_id/aa_freqs.rb +7 -3
- data/lib/spec_id/bioworks.rb +20 -14
- data/lib/spec_id/digestor.rb +139 -0
- data/lib/spec_id/mass.rb +116 -0
- data/lib/spec_id/parser/proph.rb +236 -0
- data/lib/spec_id/precision/filter/cmdline.rb +209 -0
- data/lib/spec_id/precision/filter/interactive.rb +134 -0
- data/lib/spec_id/precision/filter/output.rb +147 -0
- data/lib/spec_id/precision/filter.rb +623 -0
- data/lib/spec_id/precision/output.rb +60 -0
- data/lib/spec_id/precision/prob/cmdline.rb +139 -0
- data/lib/spec_id/precision/prob/output.rb +88 -0
- data/lib/spec_id/precision/prob.rb +171 -0
- data/lib/spec_id/proph/pep_summary.rb +92 -0
- data/lib/spec_id/proph/prot_summary.rb +484 -0
- data/lib/spec_id/proph.rb +2 -466
- data/lib/spec_id/protein_summary.rb +2 -2
- data/lib/spec_id/sequest/params.rb +316 -0
- data/lib/spec_id/sequest/pepxml.rb +1513 -0
- data/lib/spec_id/sequest.rb +2 -1672
- data/lib/spec_id/srf.rb +445 -177
- data/lib/spec_id.rb +183 -95
- data/lib/spec_id_xml.rb +8 -10
- data/lib/transmem/phobius.rb +147 -0
- data/lib/transmem/toppred.rb +368 -0
- data/lib/transmem.rb +157 -0
- data/lib/validator/aa.rb +135 -0
- data/lib/validator/background.rb +73 -0
- data/lib/validator/bias.rb +95 -0
- data/lib/validator/cmdline.rb +260 -0
- data/lib/validator/decoy.rb +94 -0
- data/lib/validator/digestion_based.rb +69 -0
- data/lib/validator/probability.rb +48 -0
- data/lib/validator/prot_from_pep.rb +234 -0
- data/lib/validator/transmem.rb +272 -0
- data/lib/validator/true_pos.rb +46 -0
- data/lib/validator.rb +214 -0
- data/lib/xml.rb +38 -0
- data/lib/xml_style_parser.rb +105 -0
- data/lib/xmlparser_wrapper.rb +19 -0
- data/script/compile_and_plot_smriti_final.rb +97 -0
- data/script/extract_gradient_programs.rb +56 -0
- data/script/get_apex_values_rexml.rb +44 -0
- data/script/mzXML2timeIndex.rb +1 -1
- data/script/smriti_final_analysis.rb +103 -0
- data/script/toppred_to_yaml.rb +47 -0
- data/script/tpp_installer.rb +1 -1
- data/{test/tc_align.rb → specs/align_spec.rb} +21 -27
- data/{test/tc_bioworks_to_pepxml.rb → specs/bin/bioworks_to_pepxml_spec.rb} +25 -41
- data/specs/bin/fasta_shaker_spec.rb +259 -0
- data/specs/bin/filter_and_validate__multiple_vals_helper.yaml +202 -0
- data/specs/bin/filter_and_validate_spec.rb +124 -0
- data/specs/bin/ms_to_lmat_spec.rb +34 -0
- data/specs/bin/prob_validate_spec.rb +62 -0
- data/specs/bin/protein_summary_spec.rb +10 -0
- data/{test/tc_fasta.rb → specs/fasta_spec.rb} +354 -310
- data/specs/gi_spec.rb +22 -0
- data/specs/load_bin_path.rb +7 -0
- data/specs/merge_deep_spec.rb +13 -0
- data/specs/ms/gradient_program_spec.rb +77 -0
- data/specs/ms/msrun_spec.rb +455 -0
- data/specs/ms/parser_spec.rb +92 -0
- data/specs/ms/spectrum_spec.rb +89 -0
- data/specs/roc_spec.rb +251 -0
- data/specs/rspec_autotest.rb +149 -0
- data/specs/sample_enzyme_spec.rb +41 -0
- data/specs/spec_helper.rb +133 -0
- data/specs/spec_id/aa_freqs_spec.rb +52 -0
- data/{test/tc_bioworks.rb → specs/spec_id/bioworks_spec.rb} +56 -71
- data/specs/spec_id/digestor_spec.rb +75 -0
- data/specs/spec_id/precision/filter/cmdline_spec.rb +20 -0
- data/specs/spec_id/precision/filter/output_spec.rb +31 -0
- data/specs/spec_id/precision/filter_spec.rb +243 -0
- data/specs/spec_id/precision/prob_spec.rb +111 -0
- data/specs/spec_id/precision/prob_spec_helper.rb +0 -0
- data/specs/spec_id/proph/pep_summary_spec.rb +143 -0
- data/{test/tc_proph.rb → specs/spec_id/proph/prot_summary_spec.rb} +52 -32
- data/{test/tc_protein_summary.rb → specs/spec_id/protein_summary_spec.rb} +85 -0
- data/specs/spec_id/sequest/params_spec.rb +68 -0
- data/specs/spec_id/sequest/pepxml_spec.rb +452 -0
- data/specs/spec_id/sqt_spec.rb +138 -0
- data/specs/spec_id/srf_spec.rb +209 -0
- data/specs/spec_id/srf_spec_helper.rb +302 -0
- data/specs/spec_id_helper.rb +33 -0
- data/specs/spec_id_spec.rb +361 -0
- data/specs/spec_id_xml_spec.rb +33 -0
- data/specs/transmem/phobius_spec.rb +423 -0
- data/specs/transmem/toppred_spec.rb +297 -0
- data/specs/transmem_spec.rb +60 -0
- data/specs/transmem_spec_shared.rb +64 -0
- data/specs/validator/aa_spec.rb +107 -0
- data/specs/validator/background_spec.rb +51 -0
- data/specs/validator/bias_spec.rb +146 -0
- data/specs/validator/decoy_spec.rb +51 -0
- data/specs/validator/fasta_helper.rb +26 -0
- data/specs/validator/prot_from_pep_spec.rb +141 -0
- data/specs/validator/transmem_spec.rb +145 -0
- data/specs/validator/true_pos_spec.rb +58 -0
- data/specs/validator_helper.rb +33 -0
- data/specs/xml_spec.rb +12 -0
- data/test_files/000_pepxml18_small.xml +206 -0
- data/test_files/020a.mzXML.timeIndex +4710 -0
- data/test_files/4-03-03_mzXML/000.mzXML.timeIndex +3973 -0
- data/test_files/4-03-03_mzXML/020.mzXML.timeIndex +3872 -0
- data/test_files/4-03-03_small-prot.xml +321 -0
- data/test_files/4-03-03_small.xml +3876 -0
- data/test_files/7MIX_STD_110802_1.sequest_params_fragment.srf +0 -0
- data/test_files/bioworks-3.3_10prots.xml +5999 -0
- data/test_files/bioworks31.params +77 -0
- data/test_files/bioworks32.params +62 -0
- data/test_files/bioworks33.params +63 -0
- data/test_files/bioworks_single_run_small.xml +7237 -0
- data/test_files/bioworks_small.fasta +212 -0
- data/test_files/bioworks_small.params +63 -0
- data/test_files/bioworks_small.phobius +109 -0
- data/test_files/bioworks_small.toppred.out +2847 -0
- data/test_files/bioworks_small.xml +5610 -0
- data/test_files/bioworks_with_INV_small.xml +3753 -0
- data/test_files/bioworks_with_SHUFF_small.xml +2503 -0
- data/test_files/corrupted_900.srf +0 -0
- data/test_files/head_of_7MIX.srf +0 -0
- data/test_files/interact-opd1_mods_small-prot.xml +304 -0
- data/test_files/messups.fasta +297 -0
- data/test_files/opd1/000.my_answer.100lines.xml +101 -0
- data/test_files/opd1/000.tpp_1.2.3.first10.xml +115 -0
- data/test_files/opd1/000.tpp_2.9.2.first10.xml +126 -0
- data/test_files/opd1/000.v2.1.mzXML.timeIndex +3748 -0
- data/test_files/opd1/000_020-prot.png +0 -0
- data/test_files/opd1/000_020_3prots-prot.mod_initprob.xml +62 -0
- data/test_files/opd1/000_020_3prots-prot.xml +62 -0
- data/test_files/opd1/opd1_cat_inv_small-prot.xml +139 -0
- data/test_files/opd1/sequest.3.1.params +77 -0
- data/test_files/opd1/sequest.3.2.params +62 -0
- data/test_files/opd1/twenty_scans.mzXML +418 -0
- data/test_files/opd1/twenty_scans.v2.1.mzXML +382 -0
- data/test_files/opd1/twenty_scans_answ.lmat +0 -0
- data/test_files/opd1/twenty_scans_answ.lmata +9 -0
- data/test_files/opd1_020_beginning.RAW +0 -0
- data/test_files/opd1_2runs_2mods/interact-opd1_mods__small.xml +753 -0
- data/test_files/orbitrap_mzData/000_cut.xml +1920 -0
- data/test_files/pepproph_small.xml +4691 -0
- data/test_files/phobius.small.noheader.txt +50 -0
- data/test_files/phobius.small.small.txt +53 -0
- data/test_files/s01_anC1_ld020mM.key.txt +25 -0
- data/test_files/s01_anC1_ld020mM.meth +0 -0
- data/test_files/small.fasta +297 -0
- data/test_files/smallraw.RAW +0 -0
- data/test_files/tf_bioworks2excel.bioXML +14340 -0
- data/test_files/tf_bioworks2excel.txt.actual +1035 -0
- data/test_files/toppred.small.out +416 -0
- data/test_files/toppred.xml.out +318 -0
- data/test_files/validator_hits_separate/bias_bioworks_small_HS.fasta +7 -0
- data/test_files/validator_hits_separate/bioworks_small_HS.xml +5651 -0
- data/test_files/yeast_gly_small-prot.xml +265 -0
- data/test_files/yeast_gly_small.1.0_1.0_1.0.parentTimes +6 -0
- data/test_files/yeast_gly_small.xml +3807 -0
- data/test_files/yeast_gly_small2.parentTimes +6 -0
- metadata +273 -57
- data/bin/filter.rb +0 -6
- data/bin/precision.rb +0 -5
- data/lib/spec/mzdata/parser.rb +0 -108
- data/lib/spec/mzdata.rb +0 -48
- data/lib/spec/mzxml/parser.rb +0 -449
- data/lib/spec/scan.rb +0 -55
- data/lib/spec_id/filter.rb +0 -797
- data/lib/spec_id/precision.rb +0 -421
- data/lib/toppred.rb +0 -18
- data/script/filter-peps.rb +0 -164
- data/test/tc_aa_freqs.rb +0 -59
- data/test/tc_fasta_shaker.rb +0 -149
- data/test/tc_filter.rb +0 -203
- data/test/tc_filter_peps.rb +0 -46
- data/test/tc_gi.rb +0 -17
- data/test/tc_id_class_anal.rb +0 -70
- data/test/tc_id_precision.rb +0 -89
- data/test/tc_msrun.rb +0 -88
- data/test/tc_mzxml.rb +0 -88
- data/test/tc_mzxml_to_lmat.rb +0 -36
- data/test/tc_peptide_parent_times.rb +0 -27
- data/test/tc_precision.rb +0 -60
- data/test/tc_roc.rb +0 -166
- data/test/tc_sample_enzyme.rb +0 -32
- data/test/tc_scan.rb +0 -26
- data/test/tc_sequest.rb +0 -336
- data/test/tc_spec.rb +0 -78
- data/test/tc_spec_id.rb +0 -201
- data/test/tc_spec_id_xml.rb +0 -36
- data/test/tc_srf.rb +0 -262
data/lib/spec_id/filter.rb
DELETED
@@ -1,797 +0,0 @@
|
|
1
|
-
|
2
|
-
require 'spec_id'
|
3
|
-
require 'optparse'
|
4
|
-
require 'ostruct'
|
5
|
-
require 'spec_id/aa_freqs'
|
6
|
-
require 'shuffle'
|
7
|
-
require 'vec'
|
8
|
-
require 'table'
|
9
|
-
|
10
|
-
|
11
|
-
########################################################
|
12
|
-
WRITE_CYS_FIND = false
|
13
|
-
########################################################
|
14
|
-
|
15
|
-
|
16
|
-
module SpecID
|
17
|
-
attr_accessor :orig_peps, :passed_peps, :passed_prots
|
18
|
-
# The filename passed in for filtering
|
19
|
-
attr_accessor :passed_in_filename
|
20
|
-
|
21
|
-
# returns the top peptide hits per file dta (first_scan + charge)
|
22
|
-
# all hits with same score as top score are returned
|
23
|
-
# assumes that all fields are strings...
|
24
|
-
# converts xcorr, deltacn, deltamass, mass, and charge into numerical types
|
25
|
-
# deletes the protein array (but not relevant proteins)
|
26
|
-
# hashes on [pep.basename, pep.first_scan.to_i, pep.charge.to_i]
|
27
|
-
# sets the @orig_peps attribute to those passing
|
28
|
-
def top_peps_prefilter!
|
29
|
-
## Bioworks peps are text based and need to be transformed first
|
30
|
-
if peps.first.is_a? Bioworks::Pep
|
31
|
-
peps.each do |pep|
|
32
|
-
pep.xcorr = pep.xcorr.to_f
|
33
|
-
pep.deltacn = pep.deltacn.to_f
|
34
|
-
pep.deltamass = pep.deltamass.to_f
|
35
|
-
pep.mass = pep.mass.to_f
|
36
|
-
pep.charge = pep.charge.to_i
|
37
|
-
pep.first_scan = pep.first_scan.to_i
|
38
|
-
end
|
39
|
-
end
|
40
|
-
## Srf Peps need no transformation!
|
41
|
-
|
42
|
-
# get the top peptide by firstscan/charge (equivalent to .out files)
|
43
|
-
top_peps = []
|
44
|
-
self.peps.hash_by {|pep| [pep.base_name, pep.first_scan, pep.charge]}.values.map do |v|
|
45
|
-
#self.peps.hash_by {|pep| [pep.aaseq, pep.charge]}.values.map do |v|
|
46
|
-
best_to_worst = v.sort_by {|pep| pep.xcorr}.reverse
|
47
|
-
top_score = best_to_worst.first.xcorr
|
48
|
-
best_to_worst.each do |pep|
|
49
|
-
if pep.xcorr == top_score
|
50
|
-
top_peps << pep
|
51
|
-
else ; break
|
52
|
-
end
|
53
|
-
end
|
54
|
-
end
|
55
|
-
@orig_peps = top_peps
|
56
|
-
end
|
57
|
-
|
58
|
-
# (xcorr1, xcorr2, xcorr3, deltacn, ppm)
|
59
|
-
# interface very unstable. For now, keeping it very loose...
|
60
|
-
# assumed that peptide xcorr, deltacn, deltamass, mass, ppm are Floats
|
61
|
-
# assumed that peptide charge is Integer
|
62
|
-
# returns peps_passed
|
63
|
-
# must respond to 'peps'
|
64
|
-
# DOES NOT UPDATE the prot.peps attribute!!
|
65
|
-
def filter_sequest(args, include_deltacnstar=false)
|
66
|
-
(x1, x2, x3, deltacn, ppm) = args
|
67
|
-
self.peps.select do |pep|
|
68
|
-
# have to add the upper limit to deltacn because the lowest score is often
|
69
|
-
# assigned a 1.10 in bioworks!
|
70
|
-
pep_deltacn = pep.deltacn
|
71
|
-
pep_charge = pep.charge
|
72
|
-
|
73
|
-
## The outer parentheses are critical to getting the correct answer!
|
74
|
-
passing = ( (pep_deltacn >= deltacn) and ((pep_charge == 1 && pep.xcorr >= x1) or (pep_charge == 2 && pep.xcorr >= x2) or (pep_charge == 3 && pep.xcorr >= x3)) and ( pep.ppm <= ppm ))
|
75
|
-
|
76
|
-
if passing
|
77
|
-
if !include_deltacnstar && pep_deltacn > 1.0
|
78
|
-
false
|
79
|
-
else
|
80
|
-
true
|
81
|
-
end
|
82
|
-
else
|
83
|
-
false
|
84
|
-
end
|
85
|
-
end
|
86
|
-
end
|
87
|
-
|
88
|
-
|
89
|
-
# given some list of SpecID::Pep based objects, finds the list of proteins
|
90
|
-
# associated with those peptides
|
91
|
-
# update_prot_peps => when true, updates prot.peps attribute given the list
|
92
|
-
# of pephits
|
93
|
-
# kind =
|
94
|
-
# :no_update (current proteins are returned, but their peps attribute
|
95
|
-
# is not updated)
|
96
|
-
# :update (current proteins returned with peps attribute updated)
|
97
|
-
# :new (new proteins are created complete with peps attribute)
|
98
|
-
def self.passing_proteins(pephits, kind=:no_update)
|
99
|
-
|
100
|
-
orig_pephits_prts = []
|
101
|
-
if kind == :new
|
102
|
-
new_prots = {}
|
103
|
-
pephits.each_with_index do |pep,i|
|
104
|
-
orig_pephits_prts[i] = pep.prots
|
105
|
-
peps_new_prts = pep.prots.map do |prt|
|
106
|
-
if new_prots.key? prt.reference
|
107
|
-
already_exists = new_prots[prt.reference]
|
108
|
-
else
|
109
|
-
np = prt.dup
|
110
|
-
np.peps = []
|
111
|
-
new_prots[np.reference] = np
|
112
|
-
np
|
113
|
-
end
|
114
|
-
end
|
115
|
-
pep.prots = peps_new_prts
|
116
|
-
end
|
117
|
-
end
|
118
|
-
|
119
|
-
if kind == :update
|
120
|
-
pephits.each do |pep|
|
121
|
-
pep.prots.each do |prt|
|
122
|
-
prt.peps = []
|
123
|
-
end
|
124
|
-
end
|
125
|
-
end
|
126
|
-
|
127
|
-
prot_set = {}
|
128
|
-
pephits.each do |pep|
|
129
|
-
prts = pep.prots
|
130
|
-
prts.each do |prt|
|
131
|
-
prot_set[ prt.reference ] = prt
|
132
|
-
end
|
133
|
-
if (kind == :update || kind == :new)
|
134
|
-
prts.each do |prt|
|
135
|
-
prt.peps << pep
|
136
|
-
end
|
137
|
-
end
|
138
|
-
end
|
139
|
-
|
140
|
-
## Reset the original protein hits
|
141
|
-
if kind == :new
|
142
|
-
pephits.each_with_index do |pep,i|
|
143
|
-
pep.prots = orig_pephits_prts[i]
|
144
|
-
end
|
145
|
-
end
|
146
|
-
|
147
|
-
prot_set.values
|
148
|
-
end
|
149
|
-
end
|
150
|
-
|
151
|
-
|
152
|
-
class SpecID::Filter
|
153
|
-
|
154
|
-
NUM_PROT_FPPR_ITERATIONS = 10
|
155
|
-
|
156
|
-
def self.run_from_argv(argv)
|
157
|
-
obj = self.new
|
158
|
-
obj.run_from_argv(argv)
|
159
|
-
end
|
160
|
-
|
161
|
-
def run_from_argv(argv)
|
162
|
-
reply = get_options(argv)
|
163
|
-
return unless reply
|
164
|
-
files, opt = reply
|
165
|
-
|
166
|
-
#files = ARGV.map {|file| file }
|
167
|
-
#ARGV.clear
|
168
|
-
|
169
|
-
$stderr.puts "reading files (can take a minute or two for large files)..." if $VERBOSE
|
170
|
-
spec_ids = files.map do |file|
|
171
|
-
spec_id = file_to_prefiltered_spec_id(file, opt)
|
172
|
-
spec_id
|
173
|
-
end
|
174
|
-
|
175
|
-
## the options hash
|
176
|
-
hash = {}
|
177
|
-
if opt.cys
|
178
|
-
if opt.cys[1]
|
179
|
-
opt.cys[1] = opt.cys[1].to_f
|
180
|
-
else
|
181
|
-
opt.cys[1] = 0.0
|
182
|
-
end
|
183
|
-
hash[:cys] = opt.cys
|
184
|
-
end
|
185
|
-
|
186
|
-
|
187
|
-
hash[:tps] =
|
188
|
-
if opt.tps
|
189
|
-
Fasta.new.read_file(opt.tps).prots.map do |prot|
|
190
|
-
prot.aaseq.chomp
|
191
|
-
end
|
192
|
-
end
|
193
|
-
|
194
|
-
hash[:dcy] =
|
195
|
-
if opt.false
|
196
|
-
new_spec_ids = []
|
197
|
-
prefixes_or_files = SpecID.extend_args(opt.false, files.size)
|
198
|
-
false_spec_ids = spec_ids.zip(prefixes_or_files).map do |spec_id, prefix_or_file|
|
199
|
-
if File.exist? prefix_or_file
|
200
|
-
new_spec_ids << spec_id
|
201
|
-
file_to_prefiltered_spec_id(prefix_or_file, opt)
|
202
|
-
else
|
203
|
-
(tps, fps) = spec_id.classify_by_false_flag(:peps, prefix_or_file, true, opt.prefix)
|
204
|
-
fps_specid = spec_id.class.new
|
205
|
-
tps_specid = spec_id.class.new
|
206
|
-
|
207
|
-
fps_specid.peps = fps
|
208
|
-
tps_specid.peps = tps
|
209
|
-
new_spec_ids << tps_specid
|
210
|
-
fps_specid
|
211
|
-
end
|
212
|
-
end
|
213
|
-
spec_ids = new_spec_ids
|
214
|
-
false_spec_ids
|
215
|
-
end
|
216
|
-
|
217
|
-
defaults = {
|
218
|
-
:dcy => nil, # { spec_id => false_spec_id }
|
219
|
-
:cys => nil, # [cys_background_freq, cys_containing_freq]
|
220
|
-
:tps => nil,
|
221
|
-
:tmm => nil,
|
222
|
-
:occams_razor => opt.occams_razor,
|
223
|
-
}
|
224
|
-
args = defaults.merge hash
|
225
|
-
|
226
|
-
|
227
|
-
base_args = [opt.x1, opt.x2, opt.x3, opt.c, opt.ppm]
|
228
|
-
|
229
|
-
#################################################### <--
|
230
|
-
@fppr_methods = [:tmm, :tps, :cys, :dcy].select do |x|
|
231
|
-
args[x]
|
232
|
-
end
|
233
|
-
@groups_reporting = [:pephits, :aaseq, :prothits]
|
234
|
-
@groups_reporting.push( :occams_razor ) if args[:occams_razor]
|
235
|
-
|
236
|
-
@cat_labels = {
|
237
|
-
:pephits => 'pep_hits',
|
238
|
-
:prothits => 'prot_hits',
|
239
|
-
:aaseq => 'uniq_aa_hits',
|
240
|
-
:occams_razor => 'occams_prot_hits',
|
241
|
-
}
|
242
|
-
#################################################### <--
|
243
|
-
|
244
|
-
if opt.log
|
245
|
-
@logfh = File.open(opt.log, 'w')
|
246
|
-
else
|
247
|
-
@logfh = nil
|
248
|
-
end
|
249
|
-
#########################################
|
250
|
-
# PRINT FILTER LEGEND
|
251
|
-
out filter_legend(@fppr_methods)
|
252
|
-
#########################################
|
253
|
-
|
254
|
-
if opt.filters_file
|
255
|
-
lines = IO.readlines(opt.filters_file)
|
256
|
-
lines.each do |line|
|
257
|
-
line.chomp!
|
258
|
-
answer = prep_reply(line, base_args)
|
259
|
-
next if answer == false
|
260
|
-
base_args = answer
|
261
|
-
filter_round(spec_ids, base_args, args)
|
262
|
-
end
|
263
|
-
elsif opt.i
|
264
|
-
## CLEAR ARGV (since otherwise, gets reads it!)
|
265
|
-
ARGV.clear
|
266
|
-
out interactive_help
|
267
|
-
reply = "nil"
|
268
|
-
loop do
|
269
|
-
b = base_args
|
270
|
-
out "#{b[0]} #{b[1]} #{b[2]} dcn:#{b[3]} ppm:#{b[4]}"
|
271
|
-
loop do
|
272
|
-
reply = gets.chomp
|
273
|
-
answer = prep_reply(reply, base_args)
|
274
|
-
if answer == false
|
275
|
-
out interactive_help
|
276
|
-
else
|
277
|
-
base_args = answer
|
278
|
-
filter_round(spec_ids, base_args, args)
|
279
|
-
break
|
280
|
-
end
|
281
|
-
end
|
282
|
-
end
|
283
|
-
else
|
284
|
-
filter_round(spec_ids, base_args, args)
|
285
|
-
end
|
286
|
-
|
287
|
-
if opt.log
|
288
|
-
@logfh.close
|
289
|
-
end
|
290
|
-
|
291
|
-
end
|
292
|
-
|
293
|
-
def out(string)
|
294
|
-
puts string
|
295
|
-
if @logfh
|
296
|
-
@logfh.puts string
|
297
|
-
end
|
298
|
-
end
|
299
|
-
|
300
|
-
# takes a fasta file or a string ( to be cast as a float )
|
301
|
-
def get_cys_freq(arg)
|
302
|
-
if File.exist? arg
|
303
|
-
SpecID::AAFreqs.new(arg).aafreqs[:C]
|
304
|
-
else
|
305
|
-
arg.to_f
|
306
|
-
end
|
307
|
-
end
|
308
|
-
|
309
|
-
# prints shortened number for display
|
310
|
-
def short(num)
|
311
|
-
sprintf( "%.3f",num)
|
312
|
-
end
|
313
|
-
|
314
|
-
# if good arguments, returns [files_array, options]
|
315
|
-
# else prints an error argument and returns nil
|
316
|
-
def get_options(argv)
|
317
|
-
dup_argv = argv.dup
|
318
|
-
|
319
|
-
opt = OpenStruct.new
|
320
|
-
opt.x1 = 1.0
|
321
|
-
opt.x2 = 1.5
|
322
|
-
opt.x3 = 2.0
|
323
|
-
opt.c = 0.1
|
324
|
-
opt.ppm = 1000.0
|
325
|
-
opt.false = false
|
326
|
-
|
327
|
-
opts = OptionParser.new do |op|
|
328
|
-
op.banner = "usage: #{File.basename(__FILE__)} [OPTS] <bioworks.xml | bioworks.srg>"
|
329
|
-
op.separator("prints number of peptides/proteins ID'd at given thresholds")
|
330
|
-
op.separator "only top hit (by xcorr) per scan+charge is considered"
|
331
|
-
|
332
|
-
#op.separator("** 'dcn*' is the number of peptides with deltacn == 1.1")
|
333
|
-
#op.separator(" (these are peptides who are the only hit with xcorr > 0)")
|
334
|
-
op.separator ""
|
335
|
-
op.on("-1", "--xcorr1 N", Float, "xcorr for +1 charge d: #{opt.x1}") {|v| opt.x1 = v}
|
336
|
-
op.on("-2", "--xcorr2 N", Float, "xcorr for +2 charge d: #{opt.x2}") {|v| opt.x2 = v}
|
337
|
-
op.on("-3", "--xcorr3 N", Float, "xcorr for +3 charge d: #{opt.x3}") {|v| opt.x3 = v}
|
338
|
-
op.on("-c", "--deltacn N", Float, ">= deltacn d: #{opt.c}") {|v| opt.c = v}
|
339
|
-
op.on("-p", "--ppm N", Float, "<= ppm d: #{opt.ppm}") {|v| opt.ppm = v}
|
340
|
-
op.separator " if bioworks.xml, = 10^6deltamass/mass"
|
341
|
-
op.on("-i", "--interactive", "interactive filtering") {|v| opt.i = v}
|
342
|
-
op.on("-f", "--false a,b,c", Array, "flag for false proteins or filenames of decoys") {|v| opt.false = v}
|
343
|
-
op.separator(" e.g., for Bioworks: 'REVERSE'")
|
344
|
-
op.separator(" (last given will apply to remaining files)")
|
345
|
-
op.on("--prefix", "match false flag for prefixes only") {|v| opt.prefix = v}
|
346
|
-
op.on("-y", "--cys <fasta_file|freq,[bkg]>", Array, "report fpr by expected cysteine freq") do |v|
|
347
|
-
v[0] = get_cys_freq(v[0])
|
348
|
-
opt.cys = v
|
349
|
-
end
|
350
|
-
op.separator(" freq = freq of cysteine as amino acid")
|
351
|
-
op.separator(" [bkg] = freq of cys containing peps d: 0.0")
|
352
|
-
op.on("--filters_file <file>", "(no -i) file with list of interactive input") {|v| opt.filters_file = v}
|
353
|
-
op.on("-t", "--tps <fasta>", "fasta file containing true hits") {|v| opt.tps = v }
|
354
|
-
#op.on("--tmm <toppred.out>", "toppred.out file with transmembr. topology") {|v| opt.tps = v }
|
355
|
-
op.on("--yaml", "spits out yaml-ized data") {|v| opt.tabulate = v }
|
356
|
-
op.on("--combined_score", "shows the combined score") {|v| opt.combined_score = v }
|
357
|
-
op.on("--marshal", "will write marshaled data or read existing") {|v| opt.marshal = v }
|
358
|
-
op.on("--log <file>", "also writes all output to file") {|v| opt.log = v }
|
359
|
-
## NEED TO IMPLEMENT THIS:
|
360
|
-
#op.on("--protein_summary", "writes passing proteins to .summary.html files") {|v| opt.protein_summary = v }
|
361
|
-
op.on("-z", "--occams_razor", "will show minimal set of proteins") {|v| opt.occams_razor = v }
|
362
|
-
end
|
363
|
-
|
364
|
-
opts.parse!(dup_argv)
|
365
|
-
|
366
|
-
if dup_argv.size < 1
|
367
|
-
puts opts
|
368
|
-
return nil
|
369
|
-
end
|
370
|
-
|
371
|
-
[dup_argv, opt]
|
372
|
-
end
|
373
|
-
|
374
|
-
# (actual # with cys, expected # with cys, total#peptides,
|
375
|
-
# mean_fraction_of_cysteines_true, std)
|
376
|
-
# PepHit(C) = Peptide containing cysteine
|
377
|
-
# # Total PepHit(C) # Observed Bad Pep (C)
|
378
|
-
# ------------------ proportional_to ----------------------
|
379
|
-
# # Total PepHit # Total Bad PepHit (X)
|
380
|
-
# returns the fppr and the total number false
|
381
|
-
def fppr_by_cysteines(ac_num_with_cys, exp_num_with_cys, total_peptides, mean_fraction_true_cys=nil, std_fraction_true_cys=nil)
|
382
|
-
|
383
|
-
# the number of bona fide BAD cysteine hits
|
384
|
-
# (some of the cysteine hits (~5%) are true positives)
|
385
|
-
|
386
|
-
ac_num_with_cys -= exp_num_with_cys * mean_fraction_true_cys if mean_fraction_true_cys
|
387
|
-
if ac_num_with_cys < 0.0 ; ac_num_with_cys = 0.0 end
|
388
|
-
total_number_false = (ac_num_with_cys * total_peptides).to_f/exp_num_with_cys
|
389
|
-
fppr = total_number_false / total_peptides
|
390
|
-
[fppr, total_number_false]
|
391
|
-
end
|
392
|
-
|
393
|
-
# num_peps_per_protein is an array of the number of peptides per protein hit
|
394
|
-
# (these are the true hits)
|
395
|
-
# assumes that the number follows a gaussian distribution (binomial
|
396
|
-
# distributions tend toward gaussians, I believe, at large N)
|
397
|
-
# returns [mean_num_wrong, mean_fppr, stdev_num_wrong, stdev_fppr] fppr
|
398
|
-
def protein_fppr( num_peps_per_protein, number_false_peptides, num_iterations=10)
|
399
|
-
|
400
|
-
## Check for more false peptides than peptides in our proteins:
|
401
|
-
total_protein_peps = 0
|
402
|
-
contained = num_peps_per_protein.each do |num|
|
403
|
-
total_protein_peps += num
|
404
|
-
end
|
405
|
-
## All peptides will be wrong every time!
|
406
|
-
## which means all proteins will be wrong every time!
|
407
|
-
if number_false_peptides >= total_protein_peps
|
408
|
-
# [all proteins wrong, fppr=1.0
|
409
|
-
return [num_peps_per_protein.size, 1.0, 0.0, 0.0]
|
410
|
-
end
|
411
|
-
|
412
|
-
|
413
|
-
num_prots = num_peps_per_protein.size
|
414
|
-
sample = VecD.new(num_iterations)
|
415
|
-
# indexed by peptide_number, pointing to a protein's peptide_count
|
416
|
-
# we shuffle the indices and then walk along until we are finished
|
417
|
-
# then we count how many proteins still have peptides
|
418
|
-
|
419
|
-
# we create an array to hold the peptide number for each protein, then we
|
420
|
-
# can reference the same entity when subtracting the peptides in the
|
421
|
-
# algorithm
|
422
|
-
cont_pep_num_per_prot_ars = (0...num_iterations).map do |i|
|
423
|
-
total_protein_peps = 0
|
424
|
-
contained = num_peps_per_protein.map do |num|
|
425
|
-
[num]
|
426
|
-
end
|
427
|
-
end
|
428
|
-
|
429
|
-
cont_num_by_pep_index_ars = cont_pep_num_per_prot_ars.map do |ar|
|
430
|
-
index_count = 0
|
431
|
-
pc_ar = []
|
432
|
-
ar.each do |contained_num|
|
433
|
-
contained_num.first.times do
|
434
|
-
pc_ar[index_count] = contained_num
|
435
|
-
index_count += 1
|
436
|
-
end
|
437
|
-
end
|
438
|
-
pc_ar
|
439
|
-
end
|
440
|
-
|
441
|
-
indices = (0...(cont_num_by_pep_index_ars.first.size)).map {|x| x }
|
442
|
-
|
443
|
-
|
444
|
-
(0...num_iterations).each do |i|
|
445
|
-
num_false = 0
|
446
|
-
indices.shuffle!
|
447
|
-
pc = cont_num_by_pep_index_ars[i]
|
448
|
-
number_false_peptides.times do |shuffle_index|
|
449
|
-
#big_i = indices[shuffle_index]
|
450
|
-
pc[indices[shuffle_index]][0] -= 1
|
451
|
-
end
|
452
|
-
cont_pep_num_per_prot_ars[i].each do |contained_pep_count|
|
453
|
-
if contained_pep_count.first == 0
|
454
|
-
num_false += 1
|
455
|
-
end
|
456
|
-
end
|
457
|
-
sample[i] = num_false
|
458
|
-
end
|
459
|
-
(mean_num_wrong, stdev) = sample.sample_stats
|
460
|
-
mean_fppr = mean_num_wrong / num_prots
|
461
|
-
stdev_fppr = stdev / num_prots
|
462
|
-
[mean_num_wrong, mean_fppr, stdev, stdev_fppr]
|
463
|
-
end
|
464
|
-
|
465
|
-
# returns [total_number_false, fppr, fraction_expected]
|
466
|
-
# also takes a hash of pephits keyed on :aaseq
|
467
|
-
def fraction_false_by_cysteines(pephits, cys_bg_freq, cys_containing_freq)
|
468
|
-
(ac, exp) = SpecID::AAFreqs.new.actual_and_expected_number_containing_cysteines(pephits, cys_bg_freq)
|
469
|
-
fraction_of_expected = ac.to_f/exp
|
470
|
-
|
471
|
-
(cys_fprate, total_num_false) = fppr_by_cysteines(ac, exp, pephits.size, cys_containing_freq)
|
472
|
-
[total_num_false, cys_fprate, fraction_of_expected]
|
473
|
-
end
|
474
|
-
|
475
|
-
def report_cysteines
|
476
|
-
#### UNDERWAY:::
|
477
|
-
cys_tps = pep_nums[i] - total_num_false
|
478
|
-
|
479
|
-
puts "CYSTEINE FPR: "
|
480
|
-
puts " (# peps containing >= 1 cysteines)"
|
481
|
-
puts " actual: #{ac}"
|
482
|
-
puts "fraction of expected: #{short(fraction_of_expected)}"
|
483
|
-
puts " expected # FP's: " + short(total_num_false)
|
484
|
-
puts " estimated FPR: " + short( 100.0*cys_fprate ) + " % "
|
485
|
-
|
486
|
-
puts "combined_score = x1 + x2 + x3 + 20.0*deltacn + 4000.0*(1.0/ppm)"
|
487
|
-
puts "Combined Score & FPR"
|
488
|
-
puts "#{combined_score}\t#{cys_fprate}"
|
489
|
-
puts "Combined Score & fraction of expected"
|
490
|
-
#puts "#{combined_score} #{fraction_of_expected}"
|
491
|
-
to_write_cys_find = ["WRITE_CYS_FIND:", combined_score, fraction_of_expected]
|
492
|
-
puts to_write_cys_find.join("\t") if WRITE_CYS_FIND
|
493
|
-
puts(['TABULATE:', combined_score, pep_tps, pep_fpr, cys_tps, cys_fprate, '', x1, x2, x3, deltacn, ppm].join("\t")) if opt.tabulate
|
494
|
-
|
495
|
-
end
|
496
|
-
|
497
|
-
def filter_legend(fppr_methods)
|
498
|
-
lines = []
|
499
|
-
lines << "Note: protein FPPR values are probably optimistic"
|
500
|
-
lines << "[this implementation assumes an equal likelihood that a false peptide"
|
501
|
-
lines << " comes from a protein with more hits as one with less (which is probably"
|
502
|
-
lines << " not the case)]"
|
503
|
-
lines << "* = deltacn_star = peptides with deltacn > 1.0 (no sibling hits)"
|
504
|
-
if fppr_methods.size > 0
|
505
|
-
lines << "Following are methods for determining false identification rate:"
|
506
|
-
lines << ['dcy=decoy', 'cys=cysteine', 'tps=known_true_positives'].join(" ")
|
507
|
-
## when tmm is implemented:
|
508
|
-
#lines << ['dcy=decoy', 'cys=cysteine', 'tmm=transmembrane', 'tps=known_true_positives'].join(" ")
|
509
|
-
end
|
510
|
-
lines.join("\n")
|
511
|
-
end
|
512
|
-
|
513
|
-
# does this give aafreq from a fasta file?
|
514
|
-
# freq = cysteines.aafreqs[:C]
|
515
|
-
|
516
|
-
# returns [total_number_false, fppr]
|
517
|
-
# pephits can be an array or a hash of peptides keyed on :aaseq
|
518
|
-
def fraction_false_by_true_pos(pephits, true_pos_aaseqs_ar)
|
519
|
-
if pephits.is_a? Hash
|
520
|
-
seqs = pephits.keys
|
521
|
-
else
|
522
|
-
seqs = pephits.map do |v|
|
523
|
-
v.aaseq
|
524
|
-
end
|
525
|
-
end
|
526
|
-
real_tps = 0
|
527
|
-
real_fps = 0
|
528
|
-
# could also do with partition
|
529
|
-
seqs.each do |pep_aaseq|
|
530
|
-
if true_pos_aaseqs_ar.any? {|prot_aaseq| prot_aaseq.include? pep_aaseq}
|
531
|
-
real_tps += 1
|
532
|
-
else
|
533
|
-
real_fps += 1
|
534
|
-
end
|
535
|
-
end
|
536
|
-
real_fppr = real_fps.to_f/pephits.size
|
537
|
-
[real_fps, real_fppr]
|
538
|
-
end
|
539
|
-
|
540
|
-
def filter_spec_id(spec_id, filter_args, args)
|
541
|
-
results_hash = {}
|
542
|
-
# that second argument is to update protein peptides
|
543
|
-
pephits = spec_id.filter_sequest(filter_args)
|
544
|
-
|
545
|
-
results_hash[:prothits] = SpecID.passing_proteins(pephits, :no_update)
|
546
|
-
results_hash[:pephits] = pephits
|
547
|
-
results_hash[:dcn_cnt] = pephits.select{|v| v.deltacn > 1.0}.size
|
548
|
-
# be aware that this is a hash keyed by aaseq and values of arrays of
|
549
|
-
# peptides sharing the same aaseq!
|
550
|
-
results_hash[:aaseq] = pephits.hash_by(:aaseq)
|
551
|
-
results_hash
|
552
|
-
end
|
553
|
-
|
554
|
-
# returns [#FP, FPPR]
|
555
|
-
def dcy_fppr(pephits, false_pephits)
|
556
|
-
fps = false_pephits.size
|
557
|
-
[fps, fps.to_f/pephits.size]
|
558
|
-
end
|
559
|
-
|
560
|
-
def tmm_fppr(pephits)
|
561
|
-
abort "NEED TO IMPLEMENT"
|
562
|
-
end
|
563
|
-
|
564
|
-
# returns [#FP, FPPR]
|
565
|
-
def cys_fppr(pephits, cys_bg_freq, cys_containing_freq)
|
566
|
-
(total_num_false, cys_fprate, fraction_of_expected) = fraction_false_by_cysteines(pephits, cys_bg_freq, cys_containing_freq)
|
567
|
-
[total_num_false, cys_fprate]
|
568
|
-
end
|
569
|
-
|
570
|
-
def tps_fppr(pephits, true_pos_aaseqs_ar)
|
571
|
-
fraction_false_by_true_pos(pephits, true_pos_aaseqs_ar)
|
572
|
-
end
|
573
|
-
|
574
|
-
## methods should be passed in like this 'cysteine' for cysteine_fppr
|
575
|
-
## all methods should return [number_false, fppr]
|
576
|
-
## returns a hash (by method) for each set of pephits
|
577
|
-
## if :dcy is given as a method, then expects the false pephits array
|
578
|
-
def calculate_pep_fppr(pephits_ar, methods, args, false_pephits_ar=nil)
|
579
|
-
cnt = 0
|
580
|
-
pephits_ar.map do |ph|
|
581
|
-
hash = {}
|
582
|
-
methods.each do |mth|
|
583
|
-
case mth
|
584
|
-
when :dcy
|
585
|
-
hash[mth.to_sym] = send("#{mth}_fppr".to_sym, ph, false_pephits_ar[cnt])
|
586
|
-
when :cys
|
587
|
-
hash[mth.to_sym] = send("#{mth}_fppr".to_sym, ph, *(args[:cys]) )
|
588
|
-
when :tps
|
589
|
-
hash[mth.to_sym] = send("#{mth}_fppr".to_sym, ph, (args[:tps]) )
|
590
|
-
else
|
591
|
-
hash[mth.to_sym] = send("#{mth}_fppr".to_sym, ph)
|
592
|
-
end
|
593
|
-
end
|
594
|
-
cnt += 1
|
595
|
-
hash
|
596
|
-
end
|
597
|
-
end
|
598
|
-
|
599
|
-
# fpr is a SpecID obj that is the false positives
|
600
|
-
# cysteines holds an aafreqs object or nil
|
601
|
-
def filter_round(spec_ids, filter_args, args)
|
602
|
-
|
603
|
-
# push fpr on the end for the calculations
|
604
|
-
## FILTER the NORMAL spec_id objects
|
605
|
-
little_tables = []
|
606
|
-
spec_ids.each_with_index do |spec_id, i|
|
607
|
-
normal_results = filter_spec_id(spec_id, filter_args, args)
|
608
|
-
|
609
|
-
## FILTER the FALSE objects (if given)
|
610
|
-
false_results =
|
611
|
-
if args[:dcy]
|
612
|
-
little_args_hash = args.dup
|
613
|
-
false_results = filter_spec_id(args[:dcy][i], filter_args, little_args_hash)
|
614
|
-
end
|
615
|
-
|
616
|
-
## HOW TO CALCULATE FPPR FOR EVERYTHING:
|
617
|
-
# pephits Fpephits C/Tpephits TPpephits
|
618
|
-
# uniqaa Funiqaa C/Tuniqaa TPuniqaa
|
619
|
-
# prothits ProtFPR(Fpephits, prothits) ProtFPR(C/Tpephits, prothits) ProtFPR(total-TPpephits, prothits)
|
620
|
-
# OccProthits ProtFPR(Funiqaa, OccProthits) ProtFPR(C/Tuniqaa, OccProthits) ProtFPR(total-TPuniqaa, OccProthits)
|
621
|
-
# C/T = cystein or Transmembrane method
|
622
|
-
|
623
|
-
## set up false results array
|
624
|
-
if args[:dcy]
|
625
|
-
fr_ar = [false_results[:pephits], false_results[:aaseq]]
|
626
|
-
else
|
627
|
-
fr_ar = nil
|
628
|
-
end
|
629
|
-
(pephits_fppr_results, aaseq_fppr_results) = calculate_pep_fppr([normal_results[:pephits], normal_results[:aaseq]], @fppr_methods, args, fr_ar)
|
630
|
-
|
631
|
-
## NORMAL prothits
|
632
|
-
## update prothits peptides
|
633
|
-
updated_proteins = SpecID.passing_proteins(normal_results[:pephits], :update)
|
634
|
-
pep_cnt_arr = updated_proteins.map {|v| v.peps.size }
|
635
|
-
|
636
|
-
## update occams prothits
|
637
|
-
if args[:occams_razor]
|
638
|
-
updated_occams_protein_triplets = SpecID::occams_razor(updated_proteins, true)
|
639
|
-
occams_pep_cnt_arr = updated_occams_protein_triplets.map {|v| v[1].size }
|
640
|
-
occams_prots = updated_occams_protein_triplets.map {|v| v[0] }
|
641
|
-
normal_results[:occams_razor] = occams_prots
|
642
|
-
end
|
643
|
-
|
644
|
-
## note that the original prot.peps arrays are obliterated by this.
|
645
|
-
## we would need to re-update if someone wanted these
|
646
|
-
|
647
|
-
prothits_fppr_results = {}
|
648
|
-
occams_results = {}
|
649
|
-
@fppr_methods.each do |mth|
|
650
|
-
prothits_fppr_results[mth] = protein_fppr(pep_cnt_arr, pephits_fppr_results[mth].first.ceil.to_i, NUM_PROT_FPPR_ITERATIONS)
|
651
|
-
occams_results[mth] = protein_fppr(occams_pep_cnt_arr, aaseq_fppr_results[mth].first.ceil.to_i, NUM_PROT_FPPR_ITERATIONS) if args[:occams_razor]
|
652
|
-
end
|
653
|
-
|
654
|
-
fppr_results = {
|
655
|
-
:pephits => pephits_fppr_results,
|
656
|
-
:aaseq => aaseq_fppr_results,
|
657
|
-
:prothits => prothits_fppr_results,
|
658
|
-
}
|
659
|
-
fppr_results[:occams_razor] = occams_results if args[:occams_razor]
|
660
|
-
|
661
|
-
## CHANGE ALL RESULTS INTO PERCENTAGES:
|
662
|
-
fppr_results.each do |bk,hash|
|
663
|
-
hash.each do |k,val|
|
664
|
-
hash[k][1] = 100.0 * val[1]
|
665
|
-
end
|
666
|
-
end
|
667
|
-
little_tables[i] = to_table( spec_id, args, normal_results, fppr_results, @groups_reporting, @fppr_methods, @cat_labels)
|
668
|
-
end
|
669
|
-
|
670
|
-
out filter_params_string(filter_args, @fppr_methods)
|
671
|
-
little_tables.each do |tbl|
|
672
|
-
out tbl.to_formatted_string(nil, ' ')
|
673
|
-
out "-----------------------------------------------\n"
|
674
|
-
end
|
675
|
-
#big_table(spec_ids, filter_args, args, normal_results, groups_reporting, fppr_results, cat_labels)
|
676
|
-
|
677
|
-
end
|
678
|
-
|
679
|
-
|
680
|
-
|
681
|
-
def filter_params_string(filter_args, fppr_methods)
|
682
|
-
(x1, x2, x3, deltacn, ppm) = filter_args
|
683
|
-
st = []
|
684
|
-
st << "=========================================================================="
|
685
|
-
st << " xcorr(1,2,3) >= #{x1},#{x2},#{x3} || deltacn >= #{deltacn} || ppm <= #{ppm} "
|
686
|
-
st << ''
|
687
|
-
st.join("\n")
|
688
|
-
#st = []
|
689
|
-
#st << ["xcorr(1,2,3) >= #{x1},#{x2},#{x3}", "deltacn >= #{deltacn}", "ppm <= #{ppm}"].join("\t")
|
690
|
-
#st
|
691
|
-
end
|
692
|
-
|
693
|
-
def to_table(spec_id, args, normal_results, fppr_results, groups_reporting, fppr_methods, cat_labels)
|
694
|
-
#table is in the form: { column heading => [ values ] }
|
695
|
-
|
696
|
-
title = spec_id.passed_in_filename
|
697
|
-
col_labels = ['num', *(fppr_methods.map{|v| "#{v}%" })]
|
698
|
-
|
699
|
-
row_labels = groups_reporting.map {|grp| cat_labels[grp]}
|
700
|
-
dt = groups_reporting.map do |grp|
|
701
|
-
line = [normal_results[grp].size]
|
702
|
-
fppr_methods.each do |mth|
|
703
|
-
line << fppr_results[grp][mth][1]
|
704
|
-
end
|
705
|
-
line
|
706
|
-
end
|
707
|
-
|
708
|
-
Table.new(dt, row_labels, col_labels, title)
|
709
|
-
#puts(['TABULATE:', combined_score, pep_tps, pep_fppr, real_tps, real_fppr, '', x1, x2, x3, deltacn, ppm].join("\t")) if opt.tabulate
|
710
|
-
end
|
711
|
-
|
712
|
-
def combined_score(filter_args)
|
713
|
-
(x1, x2, x3, deltacn, ppm) = filter_args
|
714
|
-
combined_score = x1 + x2 + x3 + 20.0*deltacn + 4000.0*(1.0/ppm)
|
715
|
-
end
|
716
|
-
|
717
|
-
# assumes its already chomped
|
718
|
-
# updates the 5 globals
|
719
|
-
def prep_reply(reply, base)
|
720
|
-
if reply == 'q' ; exit ; end
|
721
|
-
if reply =~ /^\s*$/
|
722
|
-
base
|
723
|
-
elsif reply
|
724
|
-
arr = reply.split(/\s+/)
|
725
|
-
to_change = []
|
726
|
-
to_change_hash = {}
|
727
|
-
arr.each do |it|
|
728
|
-
if it.include? ':'
|
729
|
-
(k,v) = it.split(':')
|
730
|
-
to_change_hash[k] = v
|
731
|
-
else
|
732
|
-
to_change << it
|
733
|
-
end
|
734
|
-
end
|
735
|
-
to_change.each_with_index do |tc,i|
|
736
|
-
begin
|
737
|
-
base[i] = tc.to_f
|
738
|
-
rescue NoMethodError
|
739
|
-
out "BAD ARG: #{tc}"
|
740
|
-
return false
|
741
|
-
end
|
742
|
-
end
|
743
|
-
to_change_hash.each do |k,v|
|
744
|
-
case k
|
745
|
-
when 'x1' ; base[0] = v
|
746
|
-
when 'x2' ; base[1] = v
|
747
|
-
when 'x3' ; base[2] = v
|
748
|
-
when 'dcn' ; base[3] = v
|
749
|
-
when 'ppm' ; base[4] = v
|
750
|
-
else
|
751
|
-
out "BAD ARG: #{k}:#{v}"
|
752
|
-
end
|
753
|
-
end
|
754
|
-
base.map {|v| v.to_f }
|
755
|
-
else
|
756
|
-
false
|
757
|
-
end
|
758
|
-
end
|
759
|
-
|
760
|
-
def file_to_prefiltered_spec_id(file, opt)
|
761
|
-
spec_id = nil
|
762
|
-
marshal_file = file + ".prefiltered.msh"
|
763
|
-
if File.exist?(marshal_file)
|
764
|
-
File.open(marshal_file) do |fh|
|
765
|
-
spec_id = Marshal.load(fh)
|
766
|
-
end
|
767
|
-
else
|
768
|
-
spec_id = SpecID.new(file)
|
769
|
-
spec_id.passed_in_filename = file
|
770
|
-
spec_id.top_peps_prefilter!
|
771
|
-
## marshal it!
|
772
|
-
if opt.marshal
|
773
|
-
File.open(marshal_file, "w") do |fh|
|
774
|
-
Marshal.dump(spec_id,fh)
|
775
|
-
end
|
776
|
-
end
|
777
|
-
end
|
778
|
-
spec_id
|
779
|
-
end
|
780
|
-
|
781
|
-
def interactive_help
|
782
|
-
string = []
|
783
|
-
string << "********************************************************"
|
784
|
-
string << "INTERACTIVE FILTERING HELP:"
|
785
|
-
string << "enter: <x1> <x2> <x3> <dcn> <ppm>"
|
786
|
-
string << "or : x1:<x1> x2:<x2> x3:<x3> dcn:<dcn> ppm:<ppm>"
|
787
|
-
string << "or : dcn:<dcn>"
|
788
|
-
string << "or : <x1> <x2> ppm:<ppm>"
|
789
|
-
string << "etc..."
|
790
|
-
string << "<enter> to (re)run current values"
|
791
|
-
string << "'q' to quit"
|
792
|
-
string << "********************************************************"
|
793
|
-
string.join("\n")
|
794
|
-
end
|
795
|
-
|
796
|
-
|
797
|
-
end
|