0
|
1 /******************************************************************************
|
|
2 ** @source gbaseinformationcontent
|
|
3 **
|
|
4 ** Calculates and graphs the sequence conservation using information content
|
|
5 **
|
|
6 ** @author Copyright (C) 2012 Hidetoshi Itaya
|
|
7 ** @version 1.0.3
|
|
8 ** @modified 2012/1/20 Hidetoshi Itaya Created!
|
|
9 ** @modified 2013/6/16 Revision 1
|
|
10 ** @modified 2015/2/7 Refactor
|
|
11 ** @@
|
|
12 **
|
|
13 ** This program is free software; you can redistribute it and/or
|
|
14 ** modify it under the terms of the GNU General Public License
|
|
15 ** as published by the Free Software Foundation; either version 2
|
|
16 ** of the License, or (at your option) any later version.
|
|
17 **
|
|
18 ** This program is distributed in the hope that it will be useful,
|
|
19 ** but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
20 ** MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
21 ** GNU General Public License for more details.
|
|
22 **
|
|
23 ** You should have received a copy of the GNU General Public License
|
|
24 ** along with this program; if not, write to the Free Software
|
|
25 ** Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
|
|
26 ******************************************************************************/
|
|
27
|
|
28 #include "emboss.h"
|
|
29 #include "glibs.h"
|
|
30
|
|
31
|
|
32
|
|
33
|
|
34 /* @prog gbaseinformationcontent **********************************************
|
|
35 **
|
|
36 ** Calculates and graphs the sequence conservation using information content
|
|
37 **
|
|
38 ******************************************************************************/
|
|
39
|
|
40 int main(int argc, char *argv[])
|
|
41 {
|
|
42 embInitPV("gbaseinformationcontent", argc, argv, "GEMBASSY", "1.0.3");
|
|
43
|
|
44 AjPSeqall seqall;
|
|
45 AjPSeq seq;
|
|
46 AjPStr inseq = NULL;
|
|
47
|
|
48 AjPStr position = 0;
|
|
49 ajint PatLen = 0;
|
|
50 ajint upstream = 0;
|
|
51 ajint downstream = 0;
|
|
52
|
|
53 AjBool accid = ajFalse;
|
|
54 AjPStr restid = NULL;
|
|
55 AjPStr seqid = NULL;
|
|
56
|
|
57 AjPStr base = NULL;
|
|
58 AjPStr url = NULL;
|
|
59
|
|
60 AjPFile tmpfile = NULL;
|
|
61 AjPStr tmpname = NULL;
|
|
62
|
|
63 AjBool plot = 0;
|
|
64 AjPFile outf = NULL;
|
|
65 AjPFilebuff buff = NULL;
|
|
66 AjPGraph mult = NULL;
|
|
67
|
|
68 gPlotParams gpp;
|
|
69 AjPStr title = NULL;
|
|
70
|
|
71 seqall = ajAcdGetSeqall("sequence");
|
|
72 position = ajAcdGetSelectSingle("position");
|
|
73 PatLen = ajAcdGetInt("patlen");
|
|
74 upstream = ajAcdGetInt("upstream");
|
|
75 downstream = ajAcdGetInt("downstream");
|
|
76 accid = ajAcdGetBoolean("accid");
|
|
77
|
|
78 plot = ajAcdGetToggle("plot");
|
|
79 outf = ajAcdGetOutfile("outfile");
|
|
80 mult = ajAcdGetGraphxy("graph");
|
|
81
|
|
82 base = ajStrNewC("rest.g-language.org");
|
|
83
|
|
84 gAssignUniqueName(&tmpname);
|
|
85
|
|
86 while(ajSeqallNext(seqall, &seq))
|
|
87 {
|
|
88 inseq = NULL;
|
|
89
|
|
90 if(!accid)
|
|
91 {
|
|
92 if(gFormatGenbank(seq, &inseq))
|
|
93 {
|
|
94 gAssignUniqueName(&tmpname);
|
|
95
|
|
96 tmpfile = ajFileNewOutNameS(tmpname);
|
|
97
|
|
98 if(!tmpfile)
|
|
99 {
|
|
100 ajFmtError("Output file (%S) open error\n", tmpname);
|
|
101 embExitBad();
|
|
102 }
|
|
103
|
|
104 ajFmtPrintF(tmpfile, "%S", inseq);
|
|
105 ajFileClose(&tmpfile);
|
|
106 ajFmtPrintS(&url, "http://%S/upload/upl.pl", base);
|
|
107 gFilePostSS(url, tmpname, &restid);
|
|
108 ajStrDel(&url);
|
|
109 ajSysFileUnlinkS(tmpname);
|
|
110 }
|
|
111 else
|
|
112 {
|
|
113 ajWarn("Sequence does not have features\n"
|
|
114 "Proceeding with sequence accession ID\n");
|
|
115 accid = ajTrue;
|
|
116 }
|
|
117 }
|
|
118
|
|
119 ajStrAssignS(&seqid, ajSeqGetAccS(seq));
|
|
120
|
|
121 if(ajStrGetLen(seqid) == 0)
|
|
122 {
|
|
123 ajStrAssignS(&seqid, ajSeqGetNameS(seq));
|
|
124 }
|
|
125
|
|
126 if(ajStrGetLen(seqid) == 0)
|
|
127 {
|
|
128 ajWarn("No valid header information\n");
|
|
129 }
|
|
130
|
|
131 if(accid)
|
|
132 {
|
|
133 ajStrAssignS(&restid, seqid);
|
|
134 if(ajStrGetLen(seqid) == 0)
|
|
135 {
|
|
136 ajDie("Cannot proceed without header with -accid\n");
|
|
137 }
|
|
138
|
|
139 if(!gValID(seqid))
|
|
140 {
|
|
141 ajDie("Invalid accession ID:%S, exiting\n", seqid);
|
|
142 }
|
|
143 }
|
|
144
|
|
145 url = ajStrNew();
|
|
146
|
|
147 ajFmtPrintS(&url, "http://%S/%S/base_information_content/position=%S/"
|
|
148 "PatLen=%d/upstream=%d/downstream=%d/output=f/tag=gene",
|
|
149 base, restid, position, PatLen, upstream, downstream);
|
|
150
|
|
151 if(plot)
|
|
152 {
|
|
153 title = ajStrNew();
|
|
154
|
|
155 ajStrAppendC(&title, argv[0]);
|
|
156 ajStrAppendC(&title, " of ");
|
|
157 ajStrAppendS(&title, seqid);
|
|
158
|
|
159 gpp.title = ajStrNewS(title);
|
|
160 gpp.xlab = ajStrNewC("position");
|
|
161 gpp.ylab = ajStrNewC("information content");
|
|
162
|
|
163 if(!gFilebuffURLS(url, &buff))
|
|
164 {
|
|
165 ajDie("File downloading error from:\n%S\n", url);
|
|
166 }
|
|
167
|
|
168 if(!gPlotFilebuff(buff, mult, &gpp))
|
|
169 {
|
|
170 ajDie("Error in plotting\n");
|
|
171 }
|
|
172
|
|
173 AJFREE(gpp.title);
|
|
174 AJFREE(gpp.xlab);
|
|
175 AJFREE(gpp.ylab);
|
|
176 ajStrDel(&title);
|
|
177 ajFilebuffDel(&buff);
|
|
178 }
|
|
179 else
|
|
180 {
|
|
181 ajFmtPrintF(outf, "Sequence: %S\n", seqid);
|
|
182 if(!gFileOutURLS(url, &outf))
|
|
183 {
|
|
184 ajDie("File downloading error from:\n%S\n", url);
|
|
185 }
|
|
186 }
|
|
187
|
|
188 ajStrDel(&url);
|
|
189 ajStrDel(&restid);
|
|
190 ajStrDel(&seqid);
|
|
191 ajStrDel(&inseq);
|
|
192 }
|
|
193
|
|
194 ajFileClose(&outf);
|
|
195
|
|
196 ajSeqallDel(&seqall);
|
|
197 ajSeqDel(&seq);
|
|
198 ajStrDel(&base);
|
|
199
|
|
200 ajStrDel(&position);
|
|
201
|
|
202 embExit();
|
|
203
|
|
204 return 0;
|
|
205 }
|