boolean read_codon_GeneStats(double * codon_array,char* line,FILE * ifp)
{
boolean ret = TRUE;
char * codon;
char * number;
if( strwhitestartcmp(line,"rndcodon",spacestr) != 0 ) {
warn("In reading codon line, got no 'rndcoodon' tag [%s]",line);
return FALSE;
}
while( fgets(line,MAXLINE,ifp) != NULL ) {
if( line[0] == '#' )
continue;
if( strwhitestartcmp(line,"//",spacestr) == 0 )
break;
codon = strtok(line,spacestr);
number = strtok(NULL,spacestr);
if( codon == NULL ) {
warn("Found an uncommented line in codon consensus with no leading codon word");
continue;
}
if( number == NULL ) {
warn("For codon %s, no number found",codon);
ret = FALSE;
continue;
}
if( strchr(codon,'N') != NULL )
continue;
if( is_non_ambiguous_codon_seq(codon) == FALSE ) {
warn("Codon %s is not really a codon... problem!");
ret = FALSE;
continue;
}
codon_array[base4_codon_from_seq(codon)]= atof(number);
}
return ret;
}
boolean read_base_GeneConsensus(double * base_array,char* line,FILE * ifp)
{
boolean ret = TRUE;
int b;
char * base;
char * number;
if( strwhitestartcmp(line,"begin",spacestr) != 0 || strstr(line,"consensus") == NULL ) {
warn("In reading base GeneConsensus line, got no 'begin consensus' tag [%s]",line);
return FALSE;
}
while( fgets(line,MAXLINE,ifp) != NULL ) {
if( line[0] == '#' )
continue;
if( strwhitestartcmp(line,"end",spacestr) == 0 )
break;
base = strtok(line,spacestr);
number = strtok(NULL,spacestr);
if( base == NULL ) {
warn("Found an uncommented line in base consensus with no leading base word");
continue;
}
if( number == NULL ) {
warn("For base %s, no number found",base);
ret = FALSE;
continue;
}
if( strlen(base) > 1 || (b=base_from_char(*base)) == BASE_N ) {
warn("Could not interpret %s as an actual DNA base in read_base_GeneConsensus");
ret = FALSE;
continue;
}
base_array[b]= atof(number);
}
return ret;
}
GeneConsensus * read_line_GeneConsensus(char * line,FILE * ifp)
{
GeneConsensus * out;
GeneSingleCons * temp;
char buffer[MAXLINE];
char * runner;
if( strwhitestartcmp(line,"begin",spacestr) != 0 ) {
warn("Attempting to read a GeneConsensus structure with a line not starting with 'begin' [%s]",line);
return NULL;
}
runner = strtok(line,spacestr);
runner = strtok(NULL,spacestr);
if( runner == NULL || strcmp(runner,"consensus") != 0 ) {
warn("Attempting to read a GeneConsensus structure without a 'begin consensus' tag [%s]",line);
return NULL;
}
out = GeneConsensus_alloc_std();
while( fgets(buffer,MAXLINE,ifp) != NULL ) {
if( buffer[0] == '#' )
continue;
if( strwhitestartcmp(buffer,"end",spacestr) == 0 )
break;
temp = read_line_GeneSingleCons(buffer);
if( temp == NULL ) {
warn("Unable to process GeneSingleCons line... dropping out...");
break;
}
add_GeneConsensus(out,temp);
}
return out;
}
boolean skip_consensus(FILE * ifp)
{
char buffer[MAXLINE];
while(fgets(buffer,MAXLINE,ifp) != NULL )
if( strwhitestartcmp(buffer,"end",spacestr) == 0)
break;
if( feof(ifp) || ferror(ifp) )
return FALSE;
return TRUE;
}
boolean read_into_MethodTypeSet(MethodTypeSet * mts,FILE * ifp)
{
char buffer[MAXLINE];
Method * me;
Type * ty;
Input * in;
while( fgets(buffer,MAXLINE,ifp) != NULL) {
chop_newline(buffer);
if( buffer[0] == '#' || strwhitestartcmp(buffer,"#",spacestr) == 0 )
continue;
if( only_whitespace(buffer,spacestr) == TRUE)
continue;
if( strstartcmp(buffer,"method") == 0 ) {
if( (me=read_Method_line(buffer,ifp)) == NULL ) {
warn("Unable to read method in line [%s] ",buffer);
} else {
add_me_MethodTypeSet(mts,me);
}
} else if ( strstartcmp(buffer,"type") == 0 ) {
if( (ty=read_Type_line(buffer,ifp)) == NULL ) {
warn("Unable to read type in line [%s] ",buffer);
} else {
add_ty_MethodTypeSet(mts,ty);
}
} else if ( strstartcmp(buffer,"input") == 0 ) {
if( (in = read_Input_line(buffer,ifp)) == NULL ) {
warn("Unable to read type in line [%s]",buffer);
} else {
add_in_MethodTypeSet(mts,in);
}
} else {
warn("In reading only method/types got an impossible line [%s]",buffer);
}
}
return TRUE;
}
int check_type_GeneFrequency(char *line,FILE * ifp,int * center,int * phase)
{
int ret = GeneConsensusType_Error;
char * runner;
if( strwhitestartcmp(line,"type",spacestr) != 0 ) {
warn("Attempting to check phase of consensus with no type line...");
return GeneConsensusType_Error;
}
runner = strtok(line,spacestr);
runner = strtok(NULL,spacestr);
if( runner == NULL ) {
warn("GeneFrequency type with no type. Can't read type, must set to error, but problem in later parsing");
ret = GeneConsensusType_Error;
}
else {
ret = string_to_GeneConsensusType(runner);
}
while( fgets(line,MAXLINE,ifp) != NULL ) {
if( line[0] == '#' )
continue;
else if( strwhitestartcmp(line,"phase",spacestr) == 0 ) {
runner = strtok(line,spacestr);
runner = strtok(NULL,spacestr);
if( runner == NULL ) {
warn("Got phase line with no phase. Sad....");
continue;
}
if( phase != NULL ) {
if( strcmp(runner,"all") ==0 || strcmp(runner,"All") == 0)
*phase = 3;
else *phase = atoi(runner);
}
}
else if( strwhitestartcmp(line,"center",spacestr) == 0 || strwhitestartcmp(line,"centre",spacestr) == 0) {
runner = strtok(line,spacestr);
runner = strtok(NULL,spacestr);
if( runner == NULL ) {
warn("Got center line with no phase. Sad....");
continue;
}
if( center != NULL ) {
*center = atoi(runner);
}
}
else {
break;
}
}
return ret;
}
GeneFrequency21 * read_GeneFrequency21(FILE * ifp)
{
GeneFrequency21 * out;
GeneConsensus * temp;
char buffer[MAXLINE];
int phase;
int center;
int type;
boolean err = FALSE;
out = GeneFrequency21_alloc();
while( fgets(buffer,MAXLINE,ifp) != NULL ) {
if( buffer[0] == '#' )
continue;
if( strwhitestartcmp(buffer,"type",spacestr) == 0 ) {
phase = 3; /** if no phase, assumme it is for all phases **/
type = check_type_GeneFrequency(buffer,ifp,¢er,&phase);
switch(type) {
case GeneConsensusType_5SS :
if( phase == 3) {
temp = read_line_GeneConsensus(buffer,ifp);
temp->center = center;
out->ss5 = temp;
}
else {
if( skip_consensus(ifp) == FALSE ) {
warn("Unable to skip phase'd 5'SS information ... problem!");
break;
}
}
break;
case GeneConsensusType_3SS :
if( phase == 3) {
temp = read_line_GeneConsensus(buffer,ifp);
temp->center = center;
out->ss3 = temp;
}
else {
if( skip_consensus(ifp) == FALSE ) {
warn("Unable to skip phase'd 5'SS information ... problem!");
err = TRUE;
}
}
break;
case GeneConsensusType_CDS :
if( phase == 0) {
if( read_codon_GeneConsensus(out->codon,buffer,ifp) == FALSE ) {
warn("Unable to read codon information in GeneFrequency21... problem!");
break;
}
}
else if( phase == 3 ) {
/*** we need this! ***/
if( read_codon_GeneConsensus(out->cds_triplet,buffer,ifp) == FALSE ) {
warn("Unable to read codon information in GeneFrequency21... problem!");
break;
}
}
else { /** in a different phase **/
if( skip_consensus(ifp) == FALSE ) {
warn("Unable to skip phase'd CDS information ... problem!");
err = TRUE;
}
}
break;
case GeneConsensusType_Intron_emission :
if( phase == 3 ) {
if( read_base_GeneConsensus(out->central,buffer,ifp) == FALSE ) {
warn("Unable to read Intron emissions in genefrequency21 ... problem!");
err = TRUE;
}
}
else {
if( skip_consensus(ifp) == FALSE ) {
warn("Unable to skip phase'd CDS information ... problem!");
err = TRUE;
}
}
break;
case GeneConsensusType_Pyrimidine_emission :
if( phase == 3 ) {
if( read_base_GeneConsensus(out->py,buffer,ifp) == FALSE ) {
warn("Unable to read pyrimidine emissions in genefrequency21 ... problem!");
err = TRUE;
}
}
else {
if( skip_consensus(ifp) == FALSE ) {
warn("Unable to skip phase'd pyrimidine information ... problem!");
err = TRUE;
}
}
break;
case GeneConsensusType_Spacer_emission :
if( phase == 3 ) {
if( read_base_GeneConsensus(out->spacer,buffer,ifp) == FALSE ) {
warn("Unable to read spacer emissions in genefrequency21 ... problem!");
err = TRUE;
}
}
else {
if( skip_consensus(ifp) == FALSE ) {
warn("Unable to skip phase'd spacer information ... problem!");
err = TRUE;
}
}
break;
case GeneConsensusType_Central_stay :
out->transition[GF21_CENTRAL_STAY] = double_from_line(buffer);
break;
case GeneConsensusType_Pyrimidine_stay :
out->transition[GF21_PY_STAY] = double_from_line(buffer);
break;
case GeneConsensusType_Spacer_stay :
out->transition[GF21_SPACER_STAY] = double_from_line(buffer);
break;
case GeneConsensusType_No_spacer :
out->transition[GF21_NO_SPACER] = double_from_line(buffer);
break;
case GeneConsensusType_Intron_Corr_Term :
switch(phase) {
case 0 :
/* out->transition[GF21_INTRON_CORR_TERM_0] = double_from_line(buffer); */
break;
case 1 :
/* out->transition[GF21_INTRON_CORR_TERM_1] = double_from_line(buffer); */
break;
case 2 :
/* out->transition[GF21_INTRON_CORR_TERM_2] = double_from_line(buffer); */
break;
case 3 :
out->transition[GF21_INTRON_CORR_TERM] = double_from_line(buffer);
break;
default :
warn("Well... I have got some bad news for you. We found a phase of %d in Intron correction term. ",phase);
break;
}
break;
default :
warn("Got an unidenitifable type in GeneFrequency21 parse. Skippping");
if( skip_consensus(ifp) == FALSE ) {
warn("Unable to skip phase'd 5'SS information ... problem!");
err = TRUE;
}
}
if( err == TRUE ) {
warn("You have had an unrecoverable error in GeneFrequency21 parsing");
break;
}
}
else {
striptoprint(buffer);
warn("Could not understand line [%s] in GeneFrequency21 parse",buffer);
}
}
return out;
}
