1 | #ifndef PROBE_H |
---|
2 | #define PROBE_H |
---|
3 | |
---|
4 | #ifndef __LIST__ |
---|
5 | #include <list> |
---|
6 | #endif |
---|
7 | #ifndef __SET__ |
---|
8 | #include <set> |
---|
9 | #endif |
---|
10 | |
---|
11 | #ifndef ARBDB_H |
---|
12 | #include <arbdb.h> |
---|
13 | #endif |
---|
14 | #ifndef PT_COM_H |
---|
15 | #include <PT_com.h> |
---|
16 | #endif |
---|
17 | #ifndef AISC_GEN_SERVER_INCLUDED |
---|
18 | #include <PT_server.h> |
---|
19 | #endif |
---|
20 | |
---|
21 | #define PT_SERVER_MAGIC 0x32108765 // pt server identifier |
---|
22 | #define PT_SERVER_VERSION 2 // version of pt server database (no versioning till 2009/05/13) |
---|
23 | |
---|
24 | #define pt_assert(bed) arb_assert(bed) |
---|
25 | |
---|
26 | #if defined(DEBUG) |
---|
27 | // #define PTM_DEBUG |
---|
28 | #endif // DEBUG |
---|
29 | |
---|
30 | typedef unsigned long ULONG; |
---|
31 | typedef unsigned int UINT; |
---|
32 | typedef unsigned char uchar; |
---|
33 | |
---|
34 | #define PT_CORE *(int *)0 = 0; |
---|
35 | |
---|
36 | #define PT_POS_TREE_HEIGHT 20 |
---|
37 | #define PT_POS_SECURITY 10 |
---|
38 | #define MIN_PROBE_LENGTH 6 |
---|
39 | |
---|
40 | enum PT_MATCH_TYPE { |
---|
41 | PT_MATCH_TYPE_INTEGER = 0, |
---|
42 | PT_MATCH_TYPE_WEIGHTED_PLUS_POS = 1, |
---|
43 | PT_MATCH_TYPE_WEIGHTED = -1 |
---|
44 | }; |
---|
45 | |
---|
46 | |
---|
47 | |
---|
48 | #define MATCHANSWER 50 // private msg type: match result answer |
---|
49 | #define CREATEANSWER 51 // private msg type: create result answer |
---|
50 | #define FINDANSWER 52 // private msg type: find result answer |
---|
51 | |
---|
52 | extern int gene_flag; // if 'gene_flag' == 1 -> we are a gene pt server |
---|
53 | extern ULONG physical_memory; |
---|
54 | struct Hs_struct; |
---|
55 | |
---|
56 | enum type_types { |
---|
57 | t_int = 1, |
---|
58 | t_string = 0, |
---|
59 | t_float = 2 |
---|
60 | }; |
---|
61 | |
---|
62 | enum PT_BASES { |
---|
63 | PT_QU = 0, |
---|
64 | PT_N = 1, |
---|
65 | PT_A, |
---|
66 | PT_C, |
---|
67 | PT_G, |
---|
68 | PT_T, |
---|
69 | PT_B_MAX, |
---|
70 | PT_B_UNDEF, |
---|
71 | }; |
---|
72 | |
---|
73 | // ----------------- |
---|
74 | // POS TREE |
---|
75 | |
---|
76 | enum PT_NODE_TYPE { |
---|
77 | PT_NT_LEAF = 0, |
---|
78 | PT_NT_CHAIN = 1, |
---|
79 | PT_NT_NODE = 2, |
---|
80 | PT_NT_SINGLE_NODE = 3, // stage 3 |
---|
81 | PT_NT_SAVED = 3, // stage 1+2 |
---|
82 | PT_NT_UNDEF = 4 |
---|
83 | }; |
---|
84 | |
---|
85 | struct POS_TREE { |
---|
86 | uchar flags; |
---|
87 | char data; |
---|
88 | }; |
---|
89 | |
---|
90 | struct PTM2 { |
---|
91 | char *data_start; // points to start of data |
---|
92 | int stage1; |
---|
93 | int stage2; |
---|
94 | int stage3; |
---|
95 | int mode; |
---|
96 | }; |
---|
97 | |
---|
98 | |
---|
99 | |
---|
100 | // --------------------- |
---|
101 | // Probe search |
---|
102 | |
---|
103 | class probe_input_data : virtual Noncopyable { // every taxa's own data |
---|
104 | |
---|
105 | char *data; // sequence |
---|
106 | long checksum; // checksum of sequence |
---|
107 | int size; // @@@ misleading (contains 1 if no bases in sequence) |
---|
108 | |
---|
109 | char *name; |
---|
110 | char *fullname; |
---|
111 | GBDATA *gbd; |
---|
112 | |
---|
113 | bool group; // probe_design: whether species is in group |
---|
114 | |
---|
115 | // obsolete methods below @@@ remove them |
---|
116 | GBDATA *get_gbdata() const { return gbd; } |
---|
117 | void set_data(char *assign, int size_) { pt_assert(!data); data = assign; size = size_; } |
---|
118 | void set_checksum(long cs) { checksum = cs; } |
---|
119 | |
---|
120 | public: |
---|
121 | |
---|
122 | probe_input_data() |
---|
123 | : data(0), |
---|
124 | checksum(0), |
---|
125 | size(0), |
---|
126 | name(0), |
---|
127 | fullname(0), |
---|
128 | gbd(0), |
---|
129 | group(false) |
---|
130 | {} |
---|
131 | ~probe_input_data() { |
---|
132 | free(data); |
---|
133 | free(name); |
---|
134 | free(fullname); |
---|
135 | } |
---|
136 | |
---|
137 | GB_ERROR init(GBDATA *gbd_); |
---|
138 | |
---|
139 | const char *get_data() const { return data; } |
---|
140 | char *read_alignment(int *psize) const; |
---|
141 | |
---|
142 | const char *get_name() const { return name; } |
---|
143 | const char *get_fullname() const { return fullname; } |
---|
144 | long get_checksum() const { return checksum; } |
---|
145 | int get_size() const { return size; } |
---|
146 | |
---|
147 | bool inside_group() const { return group; } |
---|
148 | bool outside_group() const { return !group; } |
---|
149 | |
---|
150 | void set_group_state(bool isGroupMember) { group = isGroupMember; } |
---|
151 | |
---|
152 | long get_abspos() const { |
---|
153 | pt_assert(gene_flag); // only legal in gene-ptserver |
---|
154 | GBDATA *gb_pos = GB_entry(get_gbdata(), "abspos"); |
---|
155 | if (gb_pos) return GB_read_int(gb_pos); |
---|
156 | return -1; |
---|
157 | } |
---|
158 | |
---|
159 | private: |
---|
160 | }; |
---|
161 | |
---|
162 | struct probe_statistic_struct { |
---|
163 | #ifdef ARB_64 |
---|
164 | int cut_offs; // statistic of chains |
---|
165 | int single_node; |
---|
166 | int short_node; |
---|
167 | int int_node; |
---|
168 | int long_node; |
---|
169 | int longs; |
---|
170 | int ints; |
---|
171 | int ints2; |
---|
172 | int shorts; |
---|
173 | int shorts2; |
---|
174 | int chars; |
---|
175 | long maxdiff; |
---|
176 | #else |
---|
177 | int cut_offs; // statistic of chains |
---|
178 | int single_node; |
---|
179 | int short_node; |
---|
180 | int long_node; |
---|
181 | int longs; |
---|
182 | int shorts; |
---|
183 | int shorts2; |
---|
184 | int chars; |
---|
185 | #endif |
---|
186 | |
---|
187 | }; |
---|
188 | |
---|
189 | class BI_ecoli_ref; |
---|
190 | |
---|
191 | class MostUsedPos : virtual Noncopyable { |
---|
192 | int pos; |
---|
193 | int used; |
---|
194 | |
---|
195 | MostUsedPos *next; |
---|
196 | |
---|
197 | void swapWith(MostUsedPos *other) { |
---|
198 | std::swap(pos, other->pos); |
---|
199 | std::swap(used, other->used); |
---|
200 | } |
---|
201 | |
---|
202 | public: |
---|
203 | MostUsedPos() : pos(0), used(0), next(NULL) { } |
---|
204 | MostUsedPos(int p) : pos(p), used(1), next(NULL) { } |
---|
205 | ~MostUsedPos() { delete next; } |
---|
206 | |
---|
207 | void clear() { |
---|
208 | pos = 0; |
---|
209 | used = 0; |
---|
210 | delete next; |
---|
211 | next = NULL; |
---|
212 | } |
---|
213 | |
---|
214 | void announce(int p) { |
---|
215 | if (p == pos) used++; |
---|
216 | else { |
---|
217 | if (next) next->announce(p); |
---|
218 | else next = new MostUsedPos(p); |
---|
219 | if (next->used>used) swapWith(next); |
---|
220 | } |
---|
221 | } |
---|
222 | |
---|
223 | |
---|
224 | int get_most_used() const { return pos; } |
---|
225 | }; |
---|
226 | |
---|
227 | extern struct probe_struct_global { |
---|
228 | GB_shell *gb_shell; |
---|
229 | GBDATA *gb_main; // ARBDB interface |
---|
230 | GBDATA *gb_species_data; |
---|
231 | GBDATA *gb_sai_data; |
---|
232 | char *alignment_name; |
---|
233 | GB_HASH *namehash; // name to int |
---|
234 | |
---|
235 | int data_count; |
---|
236 | struct probe_input_data *data; // the internal database |
---|
237 | |
---|
238 | char *ecoli; // the ecoli sequenz |
---|
239 | BI_ecoli_ref *bi_ecoli; |
---|
240 | |
---|
241 | int max_size; // maximum sequence len |
---|
242 | long char_count; // number of all 'acgtuACGTU' |
---|
243 | |
---|
244 | int mismatches; // chain handle in match |
---|
245 | double wmismatches; |
---|
246 | int N_mismatches; |
---|
247 | int w_N_mismatches[PT_POS_TREE_HEIGHT+PT_POS_SECURITY+1]; |
---|
248 | |
---|
249 | int reversed; // tell the matcher whether probe is reversed |
---|
250 | |
---|
251 | double *pos_to_weight; // position to weight |
---|
252 | char complement[256]; // complement |
---|
253 | |
---|
254 | int deep; // for probe matching |
---|
255 | int height; |
---|
256 | int length; |
---|
257 | |
---|
258 | MostUsedPos abs_pos; |
---|
259 | |
---|
260 | int sort_by; |
---|
261 | |
---|
262 | char *probe; // probe design + chains |
---|
263 | char *main_probe; |
---|
264 | |
---|
265 | char *server_name; // name of this server |
---|
266 | aisc_com *link; |
---|
267 | T_PT_MAIN main; |
---|
268 | Hs_struct *com_so; // the communication socket |
---|
269 | POS_TREE *pt; |
---|
270 | PTM2 *ptmain; |
---|
271 | |
---|
272 | probe_statistic_struct stat; |
---|
273 | |
---|
274 | } psg; |
---|
275 | |
---|
276 | class gene_struct { |
---|
277 | char *gene_name; |
---|
278 | const char *arb_species_name; // pointers into 'gene_name' |
---|
279 | const char *arb_gene_name; |
---|
280 | |
---|
281 | void init(const char *gene_name_, const char *arb_species_name_, const char *arb_gene_name_) { |
---|
282 | int gene_name_len = strlen(gene_name_); |
---|
283 | int arb_species_name_len = strlen(arb_species_name_); |
---|
284 | int arb_gene_name_len = strlen(arb_gene_name_); |
---|
285 | |
---|
286 | int fulllen = gene_name_len+1+arb_species_name_len+1+arb_gene_name_len+1; |
---|
287 | gene_name = new char[fulllen]; |
---|
288 | strcpy(gene_name, gene_name_); |
---|
289 | arb_species_name = gene_name+(gene_name_len+1); |
---|
290 | strcpy((char*)arb_species_name, arb_species_name_); |
---|
291 | arb_gene_name = arb_species_name+(arb_species_name_len+1); |
---|
292 | strcpy((char*)arb_gene_name, arb_gene_name_); |
---|
293 | } |
---|
294 | |
---|
295 | public: |
---|
296 | gene_struct(const char *gene_name_, const char *arb_species_name_, const char *arb_gene_name_) { |
---|
297 | init(gene_name_, arb_species_name_, arb_gene_name_); |
---|
298 | } |
---|
299 | gene_struct(const gene_struct& other) { |
---|
300 | if (&other != this) { |
---|
301 | init(other.get_internal_gene_name(), other.get_arb_species_name(), other.get_arb_gene_name()); |
---|
302 | } |
---|
303 | } |
---|
304 | gene_struct& operator = (const gene_struct& other) { |
---|
305 | if (&other != this) { |
---|
306 | delete [] gene_name; |
---|
307 | init(other.get_internal_gene_name(), other.get_arb_species_name(), other.get_arb_gene_name()); |
---|
308 | } |
---|
309 | return *this; |
---|
310 | } |
---|
311 | |
---|
312 | ~gene_struct() { |
---|
313 | delete [] gene_name; |
---|
314 | } |
---|
315 | |
---|
316 | const char *get_internal_gene_name() const { return gene_name; } |
---|
317 | const char *get_arb_species_name() const { return arb_species_name; } |
---|
318 | const char *get_arb_gene_name() const { return arb_gene_name; } |
---|
319 | }; |
---|
320 | |
---|
321 | struct ltByArbName { |
---|
322 | bool operator()(const gene_struct *gs1, const gene_struct *gs2) const { |
---|
323 | int cmp = strcmp(gs1->get_arb_species_name(), gs2->get_arb_species_name()); |
---|
324 | if (cmp == 0) { cmp = strcmp(gs1->get_arb_gene_name(), gs2->get_arb_gene_name()); } |
---|
325 | return cmp<0; |
---|
326 | } |
---|
327 | }; |
---|
328 | struct ltByInternalName { |
---|
329 | bool operator()(const gene_struct *gs1, const gene_struct *gs2) const { |
---|
330 | int cmp = strcmp(gs1->get_internal_gene_name(), gs2->get_internal_gene_name()); |
---|
331 | return cmp<0; |
---|
332 | } |
---|
333 | }; |
---|
334 | |
---|
335 | typedef std::list<gene_struct> gene_struct_list; |
---|
336 | typedef std::set<const gene_struct *, ltByInternalName> gene_struct_index_internal; |
---|
337 | typedef std::set<const gene_struct *, ltByArbName> gene_struct_index_arb; |
---|
338 | |
---|
339 | extern gene_struct_index_arb gene_struct_arb2internal; // sorted by arb species+gene name |
---|
340 | extern gene_struct_index_internal gene_struct_internal2arb; // sorted by internal name |
---|
341 | |
---|
342 | #define PT_base_string_counter_eof(str) (*(unsigned char *)(str) == 255) |
---|
343 | |
---|
344 | #else |
---|
345 | #error probe.h included twice |
---|
346 | #endif |
---|
347 | |
---|
348 | |
---|