Int2 Main()
{
AsnIoPtr aip;
BioseqPtr bsp;
Pointer dataptr;
Uint2 datatype;
Boolean found;
SPI_mRNAToHerdPtr h_head;
SPI_mRNAToHerdPtr h_prev;
SPI_mRNAToHerdPtr hptr;
FILE *ifp;
Boolean isGIlist;
Char line[60];
Boolean lowercase;
SeqLocPtr lcaseloc;
FILE *ofp;
FILE *ofp2;
SeqAlignPtr sap;
SeqAnnotPtr sanp;
SeqEntryPtr sep;
FILE *sfp;
SeqIdPtr sip;
SeqLocPtr slp;
SPI_bsinfoPtr spig;
SPI_bsinfoPtr spig_head;
SPI_bsinfoPtr spig_prev;
SPI_bsinfoPtr spim;
SPI_bsinfoPtr spim_head;
SPI_bsinfoPtr spim_prev;
SPI_OptionsPtr spot;
SPI_RegionInfoPtr srip = NULL;
SPI_RegionInfoPtr srip_head;
SPI_RegionInfoPtr srip_prev;
CharPtr str;
CharPtr txt;
ID1BioseqFetchEnable("spidey", FALSE);
LocalSeqFetchInit(FALSE);
/* standard setup */
ErrSetFatalLevel (SEV_MAX);
ErrClearOptFlags (EO_SHOW_USERSTR);
UseLocalAsnloadDataAndErrMsg ();
ErrPathReset ();
if (! AllObjLoad ())
{
Message (MSG_FATAL, "AllObjLoad failed");
return 1;
}
if (! SubmitAsnLoad ())
{
Message (MSG_FATAL, "SubmitAsnLoad failed");
return 1;
}
if (! FeatDefSetLoad ())
{
Message (MSG_FATAL, "FeatDefSetLoad failed");
return 1;
}
if (! SeqCodeSetLoad ())
{
Message (MSG_FATAL, "SeqCodeSetLoad failed");
return 1;
}
if (! GeneticCodeTableLoad ())
{
Message (MSG_FATAL, "GeneticCodeTableLoad failed");
return 1;
}
if (!GetArgs("SPIDEY", NUMARGS, myargs))
return 0;
/* set the error message level high to suppress warnings from BLAST */
isGIlist = (Boolean)myargs[MYARGGILIST].intvalue;
txt = myargs[MYARGGENFILE].strvalue;
ifp = FileOpen(txt, "r");
spig_head = NULL;
if (ifp == NULL)
{
bsp = SPI_GetBspFromGIOrAcc(txt);
if (bsp == NULL)
{
ErrPostEx(SEV_ERROR, 0, 0, "Can't open genomic input file\n");
return -1;
} else
{
spig_head = (SPI_bsinfoPtr)MemNew(sizeof(SPI_bsinfo));
spig_head->bsp = bsp;
}
}
if (spig_head == NULL)
{
spig_prev = NULL;
/* read in the genomic sequence(s) first and put them into bsinfo structures */
while ((dataptr = ReadAsnFastaOrFlatFile (ifp, &datatype, NULL, FALSE, FALSE, TRUE, FALSE)) != NULL)
{
if (datatype == OBJ_BIOSEQ)
{
spig = (SPI_bsinfoPtr)MemNew(sizeof(SPI_bsinfo));
spig->bsp = (BioseqPtr)dataptr;
if (spig_head == NULL)
spig_head = spig_prev = spig;
else
{
spig_prev->next = spig;
spig_prev = spig;
}
} else if (datatype == OBJ_SEQENTRY)
{
sep = (SeqEntryPtr)dataptr;
SeqEntryExplore(sep, &spig_head, SPI_FindAllNuc);
}
}
FileClose(ifp);
}
if (spig_head == NULL)
{
ErrPostEx(SEV_ERROR, 0, 0, "No valid bioseqs in genomic file\n");
return -1;
} else if (ISA_aa(spig_head->bsp->mol))
{
ErrPostEx(SEV_ERROR, 0, 0, "At least one of the genomic sequences appears to be a protein.\n");
return -1;
}
if (spig_head->next != NULL)
{
ErrPostEx(SEV_ERROR, 0, 0, "This version can only process one genomic sequence at a time. Only the first sequence in this file will be used.\n");
spig_head->next = NULL;
}
spim_head = spim_prev = NULL;
txt = myargs[MYARGMRNAFILE].strvalue;
ifp = FileOpen(txt, "r");
if (ifp == NULL)
{
bsp = SPI_GetBspFromGIOrAcc(txt);
if (bsp == NULL)
{
ErrPostEx(SEV_ERROR, 0, 0, "Can't open mRNA input file\n");
return -1;
} else
{
spim_head = (SPI_bsinfoPtr)MemNew(sizeof(SPI_bsinfo));
spim_head->bsp = bsp;
}
}
if (spim_head == NULL)
{
lowercase = (Boolean)myargs[MYARGMASKED].intvalue;
lcaseloc = NULL;
/* if the mRNA has lowercase masking, read it in carefully to record the masking */
if (lowercase == TRUE)
{
while ((sep = FastaToSeqEntryForDb(ifp, TRUE, NULL, TRUE, NULL, NULL, &lcaseloc)) != NULL)
{
SeqEntryExplore(sep, &spim_head, SPI_FindAllNuc);
if (lcaseloc != NULL) /* put masking info into the bsinfo structure */
{
spim = spim_head;
sip = SeqLocId(lcaseloc);
found = FALSE;
while (spim != NULL && !found)
{
if (SeqIdComp(sip, spim->bsp->id) == SIC_YES)
{
found = TRUE;
spim->lcaseloc = lcaseloc;
}
spim = spim->next;
}
lcaseloc = NULL;
}
}
} else if (isGIlist) /* mRNA file is a list of GIs, must fetch the bioseqs */
{
str = ReadALine(line, sizeof(line), ifp);
while (str != NULL)
{
bsp = SPI_GetBspFromGIOrAcc(str);
if (bsp != NULL)
{
spim = (SPI_bsinfoPtr)MemNew(sizeof(SPI_bsinfo));
spim->bsp = bsp;
if (spim_head == NULL)
spim_head = spim_prev = spim;
else
{
spim_prev->next = spim;
spim_prev = spim;
}
}
str = ReadALine(line, sizeof(line), ifp);
}
} else /* mRNAs are FASTA or ASN.1, read them all in */
{
while ((dataptr = ReadAsnFastaOrFlatFile (ifp, &datatype, NULL, FALSE, FALSE, TRUE, FALSE)) != NULL)
{
if (datatype == OBJ_BIOSEQ)
{
spim = (SPI_bsinfoPtr)MemNew(sizeof(SPI_bsinfo));
spim->bsp = (BioseqPtr)dataptr;
if (spim_head == NULL)
spim_head = spim_prev = spim;
else
{
spim_prev->next = spim;
spim_prev = spim;
}
} else if (datatype == OBJ_SEQENTRY)
{
sep = (SeqEntryPtr)dataptr;
SeqEntryExplore(sep, &spim_head, SPI_FindAllNuc);
}
}
}
FileClose(ifp);
}
if (spim_head == NULL)
{
ErrPostEx(SEV_ERROR, 0, 0, "No valid bioseqs in mRNA file\n");
return -1;
} else if (ISA_aa(spim_head->bsp->mol))
{
ErrPostEx(SEV_ERROR, 0, 0, "At least one of the mRNA sequences appears to be a protein\n");
return -1;
}
txt = myargs[MYARGTABLE].strvalue;
if (txt != NULL)
{
ifp = FileOpen(txt, "r");
if (ifp == NULL)
{
ErrPostEx(SEV_ERROR, 0, 0, "Unable to open table file\n");
return -1;
}
SPI_ReadFeatureTable(ifp, spim_head);
spim = spim_head;
while (spim != NULL)
{
if (spim->lcaseloc != NULL)
{
slp = (SeqLocPtr)ValNodeNew(NULL);
slp->choice = SEQLOC_MIX;
slp->data.ptrvalue = (Pointer)spim->lcaseloc;
spim->lcaseloc = slp;
}
spim = spim->next;
}
}
spim = spim_head;
spot = (SPI_OptionsPtr)MemNew(sizeof(SPI_Options));
spot->printaln = myargs[MYARGPRALIGN].intvalue;
txt = myargs[MYARGOUTFILE].strvalue;
ofp = FileOpen(txt, "w");
if (ofp == NULL && spot->printaln != 3)
{
ErrPostEx(SEV_ERROR, 0, 0, "Unable to open output file\n");
return -1;
}
if (spot->printaln >= 2)
{
txt = myargs[MYARGALNFILE].strvalue;
ofp2 = FileOpen(txt, "a");
if (ofp2 == NULL)
{
ErrPostEx(SEV_ERROR, 0, 0, "Unable to open output file 2\n");
return -1;
}
} else
ofp2 = NULL;
/** ErrSetMessageLevel(SEV_MAX); **/
spot->firstpasseval = myargs[MYARG1STEVAL].floatvalue;
spot->secpasseval = myargs[MYARG2NDEVAL].floatvalue;
spot->thirdpasseval = myargs[MYARG3RDEVAL].floatvalue;
spot->numreturns = myargs[MYARGNUMMOD].intvalue;
spot->idcutoff = myargs[MYARGIDCUT].intvalue;
spot->lencutoff = myargs[MYARGLENCUT].intvalue;
spot->interspecies = (Boolean)myargs[MYARGSPEC].intvalue;
spot->printasn = (Boolean)myargs[MYARGASN].intvalue;
spot->fetchcds = (Boolean)myargs[MYARGGETCDS].intvalue;
/*spot->ace = (Boolean)myargs[MYARGACEDB].intvalue;*/
spot->from = myargs[MYARGFROM].intvalue;
spot->to = myargs[MYARGTO].intvalue;
spot->makemult = (Boolean)myargs[MYARGMULT].intvalue;
/*KSK*/
spot->bigintron = (Boolean)myargs[MYARGXL].intvalue;
spot->bigintron_size = myargs[MYARGXL_SIZE].intvalue;
spot->repeat_db_file = myargs[MYARGREPDB].strvalue;
txt = myargs[MYARGORG].strvalue;
if (!StringICmp(txt, "d") || !StringICmp(txt, "D")){
spot->organism = SPI_FLY;
}
else if (!StringICmp(txt, "p") || !StringICmp(txt, "P")){
spot->organism = SPI_PLANT;
}
else if (!StringICmp(txt, "c") || !StringICmp(txt, "C")){
spot->organism = SPI_CELEGANS;
}
else if (!StringICmp(txt, "m") || !StringICmp(txt, "M")){
spot->organism = SPI_DICTY;
}
else {
spot->organism = SPI_VERTEBRATE;
}
sap = NULL;
if (spot->printasn)
spot->sap_head = &sap;
txt = myargs[MYARGSTRAND].strvalue;
if (txt != NULL)
{
if (StrChr(txt, 'p') || StrChr(txt, 'P'))
spot->strand = Seq_strand_plus;
else
spot->strand = Seq_strand_minus;
} else
spot->strand = Seq_strand_both;
/*txt = myargs[MYARGDRAFTFILE].strvalue;
if (txt != NULL)
spot->draftfile = StringSave(txt);*/
txt = myargs[MYARGDSPLICE].strvalue;
if (txt != NULL)
{
sfp = FileOpen(txt, "r");
SPI_GetSpliceInfo(spot, sfp, TRUE);
FileClose(sfp);
}
txt = myargs[MYARGASPLICE].strvalue;
if (txt != NULL)
{
sfp = FileOpen(txt, "r");
SPI_GetSpliceInfo(spot, sfp, FALSE);
FileClose(sfp);
}
h_head = h_prev = NULL;
srip_head = srip_prev = NULL;
while (spim != NULL)
{
spot->lcaseloc = spim->lcaseloc;
if (spot->draftfile == NULL)
srip = SPI_AlnSinglemRNAToGen(spig_head, spim, ofp, ofp2, spot);
else
{
hptr = SPI_AlnSinglemRNAToPieces(spig_head, spim, ofp, ofp2, spot);
if (h_head != NULL)
{
h_prev->next = hptr;
h_prev = hptr;
} else
h_head = h_prev = hptr;
}
if (srip != NULL)
{
if (srip_head != NULL)
{
srip_prev->next = srip;
srip_prev = srip;
} else
srip_head = srip_prev = srip;
}
spim = spim->next;
}
if (spot->makemult)
{
SPI_MakeMultipleAlignment(srip_head);
SPI_PrintMultipleAlignment(srip_head, FALSE, spig_head->bsp, ofp);
SPI_RegionListFree(srip_head);
} else
SPI_RegionListFree(srip_head);
/* create the ASN.1 output, if requested; need to use the continuous alignment */
/* that was generated */
if (spot->printasn && *(spot->sap_head) != NULL && spot->draftfile == NULL)
{
sanp = SeqAnnotForSeqAlign(*(spot->sap_head));
txt = myargs[MYARGASNFILE].strvalue;
aip = AsnIoOpen(txt, "w");
SeqAnnotAsnWrite(sanp, aip, NULL);
AsnIoClose(aip);
SeqAlignSetFree(*(spot->sap_head));
}
FileClose(ofp);
FileClose(ofp2);
SPI_OptionsFree(spot);
SPI_bsinfoFreeList(spim_head);
SPI_bsinfoFreeList(spig_head);
LocalSeqFetchDisable();
ID1BioseqFetchDisable();
return 0;
}
static Int4
BL2SEQ_GetSequences(Boolean seq1_is_na, Boolean seq2_is_na, BioseqPtr *query_bsp, BioseqPtr *subject_bsp,
SeqEntryPtr *sep, SeqEntryPtr *sep1, SeqLocPtr *lcase_mask, Boolean believe_query)
{
Boolean entrez_lookup = (Boolean) myargs[ARG_ACCN].intvalue;
char *query_accver = NULL, *subject_accver = NULL; /* Used if entrez_lookup. */
char *blast_inputfile = NULL, *blast_inputfile1 = NULL; /* Used if FASTA read. */
if (entrez_lookup) {
query_accver = myargs [ARG_QUERY].strvalue;
subject_accver = myargs [ARG_SUBJECT].strvalue;
} else {
blast_inputfile = myargs [ARG_QUERY].strvalue;
blast_inputfile1 = myargs [ARG_SUBJECT].strvalue;
}
if (entrez_lookup) {
*query_bsp = BioseqFromAccession(query_accver, seq1_is_na);
} else {
FILE *infp;
if ((infp = FileOpen(blast_inputfile, "r")) == NULL)
{
ErrPostEx(SEV_FATAL, 1, 0, "blast: Unable to open input file %s\n", blast_inputfile);
return FALSE;
}
if (myargs[ARG_LCASE].intvalue)
*sep = FastaToSeqEntryForDb(infp, seq1_is_na, NULL,
believe_query, NULL, NULL,
lcase_mask);
else
*sep = FastaToSeqEntryEx(infp, seq1_is_na, NULL, believe_query);
FileClose(infp);
if (*sep != NULL) {
*query_bsp = NULL;
if (seq1_is_na)
SeqEntryExplore(*sep, query_bsp, FindNuc);
else
SeqEntryExplore(*sep, query_bsp, FindProt);
}
}
if (*query_bsp == NULL) {
ErrPostEx(SEV_FATAL, 1, 0, "Unable to obtain bioseq\n");
return FALSE;
}
if (entrez_lookup) {
*subject_bsp =
BioseqFromAccession(subject_accver, seq2_is_na);
} else {
FILE *infp1;
if ((infp1 = FileOpen(blast_inputfile1, "r")) == NULL)
{
ErrPostEx(SEV_FATAL, 1, 0, "blast: Unable to open input file %s\n", blast_inputfile1);
return FALSE;
}
*sep1 = FastaToSeqEntryEx(infp1, seq2_is_na, NULL, FALSE);
FileClose(infp1);
if (*sep1 != NULL) {
*subject_bsp = NULL;
if (seq2_is_na)
SeqEntryExplore(*sep1, subject_bsp, FindNuc);
else
SeqEntryExplore(*sep1, subject_bsp, FindProt);
}
}
if (*subject_bsp == NULL) {
ErrPostEx(SEV_FATAL, 1, 0, "Unable to obtain bioseq\n");
return FALSE;
}
return TRUE;
}
static Int2 Main_old (void)
{
AsnIoPtr aip, xml_aip = NULL;
BioseqPtr query_bsp, PNTR query_bsp_array;
BioSourcePtr source;
BLAST_MatrixPtr matrix;
BLAST_OptionsBlkPtr options;
BLAST_KarlinBlkPtr ka_params=NULL, ka_params_gap=NULL;
BlastPruneSapStructPtr prune;
Boolean db_is_na, query_is_na, show_gi, believe_query=FALSE;
Boolean html=FALSE;
CharPtr params_buffer=NULL;
Int4 number_of_descriptions, number_of_alignments;
SeqAlignPtr seqalign, PNTR seqalign_array;
SeqAnnotPtr seqannot;
SeqEntryPtr PNTR sepp;
TxDfDbInfoPtr dbinfo=NULL, dbinfo_head;
Uint1 align_type, align_view, out_type;
Uint4 align_options, print_options;
ValNodePtr mask_loc, mask_loc_start, next_mask_loc;
ValNodePtr vnp, other_returns, error_returns;
CharPtr blast_program, blast_database, blast_inputfile, blast_outputfile;
FILE *infp, *outfp, *mqfp=NULL;
Int4 index, num_bsps, total_length, total_processed = 0;
Int2 ctr = 1;
Char prefix[2];
SeqLocPtr last_mask, mask_slp;
Boolean done, hits_found;
Boolean lcase_masking;
MBXmlPtr mbxp = NULL;
Boolean traditional_formatting;
blast_program = "blastn";
blast_database = myargs [ARG_DB].strvalue;
blast_inputfile = myargs [ARG_QUERY].strvalue;
blast_outputfile = myargs [ARG_OUT].strvalue;
if (myargs[ARG_HTML].intvalue)
html = TRUE;
if ((infp = FileOpen(blast_inputfile, "r")) == NULL) {
ErrPostEx(SEV_FATAL, 1, 0, "mgblast: Unable to open input file %s\n", blast_inputfile);
return (1);
}
align_view = (Int1) myargs[ARG_FORMAT].intvalue;
/* Geo mod:
-- replaced myargs[ARG_OUTTYPE].intvalue with out_type from now on
*/
out_type=(Int1) myargs[ARG_OUTTYPE].intvalue;
if (out_type==MGBLAST_FLTHITS || out_type==MGBLAST_HITGAPS) {
align_view = 12 + (out_type-MGBLAST_FLTHITS );
out_type=MBLAST_ALIGNMENTS;
//Attention: 12 MUST be the -m mgblast tab option for MGBLAST_FLTHITS format
// and MGBLAST_HITGAPS = MGBLAST_FLTHITS+1
if (align_view>12) { // this is MGBLAST_HITGAPS output
gap_Info=TRUE;
if (dbgaps_buf==NULL)
dbgaps_buf=(CharPtr) Malloc(dbgaps_bufsize + 1);
if (qgaps_buf==NULL)
qgaps_buf=(CharPtr) Malloc(qgaps_bufsize + 1);
}
}
outfp = NULL;
traditional_formatting =
(out_type == MBLAST_ALIGNMENTS ||
out_type == MBLAST_DELAYED_TRACEBACK);
if ((!traditional_formatting ||
(align_view != 7 && align_view != 10 && align_view != 11)) &&
blast_outputfile != NULL) {
if ((outfp = FileOpen(blast_outputfile, "w")) == NULL) {
ErrPostEx(SEV_FATAL, 1, 0, "blast: Unable to open output file %s\n", blast_outputfile);
return (1);
}
}
//align_type = BlastGetTypes(blast_program, &query_is_na, &db_is_na);
align_type = BlastGetTypes(blast_program, &query_is_na, &db_is_na);
/*
if (!traditional_formatting)
believe_query = TRUE;
else
believe_query = (Boolean) myargs[ARG_BELIEVEQUERY].intvalue;
*/
//Geo mod:
believe_query=FALSE;
//If ASN.1 output is requested and believe_query is not set to TRUE,
// exit with an error.
if (!believe_query && (myargs[ARG_ASNOUT].strvalue ||
align_view == 10 || align_view == 11)) {
ErrPostEx(SEV_FATAL, 1, 0,
"-J option must be TRUE to produce ASN.1 output; before "
"changing -J to TRUE please also ensure that all query "
"sequence identifiers are unique");
return -1;
}
options = BLASTOptionNewEx(blast_program, TRUE, TRUE);
if (options == NULL)
return 3;
options->do_sum_stats = FALSE;
options->is_neighboring = FALSE;
options->expect_value = (Nlm_FloatHi) myargs [ARG_EVALUE].floatvalue;
number_of_descriptions = myargs[ARG_DESCRIPTIONS].intvalue;
number_of_alignments = myargs[ARG_ALIGNMENTS].intvalue;
options->hitlist_size = MAX(number_of_descriptions, number_of_alignments);
if (myargs[ARG_XDROP].intvalue != 0)
options->gap_x_dropoff = myargs[ARG_XDROP].intvalue;
if (myargs[ARG_XDROP_UNGAPPED].intvalue != 0)
options->dropoff_2nd_pass = myargs[ARG_XDROP_UNGAPPED].intvalue;
if (myargs[ARG_XDROP_FINAL].intvalue != 0)
options->gap_x_dropoff_final = myargs[ARG_XDROP_FINAL].intvalue;
if (StringICmp(myargs[ARG_FILTER].strvalue, "T") == 0)
options->filter_string = StringSave("D");
else
options->filter_string = StringSave(myargs[ARG_FILTER].strvalue);
show_gi = (Boolean) myargs[ARG_SHOWGIS].intvalue;
options->penalty = myargs[ARG_MISMATCH].intvalue;
options->reward = myargs[ARG_MATCH].intvalue;
if (myargs[ARG_GAPOPEN].intvalue >= 0)
options->gap_open = myargs[ARG_GAPOPEN].intvalue;
if (myargs[ARG_GAPEXT].intvalue >= 0)
options->gap_extend = myargs[ARG_GAPEXT].intvalue;
if (options->gap_open == 0 && options->reward % 2 == 0 &&
options->gap_extend == options->reward / 2 - options->penalty)
/* This is the default value */
options->gap_extend = 0;
options->genetic_code = 1;
options->db_genetic_code = 1; /* Default; it's not needed here anyway */
options->number_of_cpus = myargs[ARG_THREADS].intvalue;
if (myargs[ARG_WORDSIZE].intvalue != 0)
options->wordsize = myargs[ARG_WORDSIZE].intvalue;
if (myargs[ARG_MINSCORE].intvalue == 0)
options->cutoff_s2 = options->wordsize*options->reward;
else
options->cutoff_s2 = myargs[ARG_MINSCORE].intvalue;
options->db_length = (Int8) myargs[ARG_DBSIZE].floatvalue;
options->searchsp_eff = (Nlm_FloatHi) myargs[ARG_SEARCHSP].floatvalue;
options->perform_culling = FALSE;
/* Kludge */
options->block_width = myargs[ARG_MAXPOS].intvalue;
options->strand_option = myargs[ARG_STRAND].intvalue;
options->window_size = myargs[ARG_WINDOW].intvalue;
#ifdef DO_NOT_SUPPRESS_BLAST_OP
options->mb_template_length = myargs[ARG_TEMPL_LEN].intvalue;
if (myargs[ARG_TEMPL_LEN].intvalue != 0)
options->mb_one_base_step = (Boolean) myargs[ARG_EVERYBASE].intvalue;
options->mb_disc_type = myargs[ARG_TEMPL_TYPE].intvalue;
#endif
lcase_masking = (Boolean) myargs[ARG_LCASE].intvalue;
/* Allow dynamic programming gapped extension only with affine
gap scores */
if (options->gap_open != 0 || options->gap_extend != 0)
options->mb_use_dyn_prog = (Boolean) myargs[ARG_DYNAMIC].intvalue;
print_options = 0;
align_options = 0;
align_options += TXALIGN_COMPRESS;
align_options += TXALIGN_END_NUM;
if (show_gi) {
align_options += TXALIGN_SHOW_GI;
print_options += TXALIGN_SHOW_GI;
}
if (align_view) {
align_options += TXALIGN_MASTER;
if (align_view == 1 || align_view == 3)
align_options += TXALIGN_MISMATCH;
if (align_view == 3 || align_view == 4 || align_view == 6)
align_options += TXALIGN_FLAT_INS;
if (align_view == 5 || align_view == 6)
align_options += TXALIGN_BLUNT_END;
} else {
align_options += TXALIGN_MATRIX_VAL;
align_options += TXALIGN_SHOW_QS;
}
if (html) {
align_options += TXALIGN_HTML;
print_options += TXALIGN_HTML;
}
if (myargs[ARG_GILIST].strvalue)
options->gifile = StringSave(myargs[ARG_GILIST].strvalue);
if (out_type == MBLAST_ENDPOINTS)
options->no_traceback = 1;
else if (out_type == MBLAST_DELAYED_TRACEBACK)
options->no_traceback = 2;
else
options->no_traceback = 0;
options->megablast_full_deflines = (Boolean) myargs[ARG_FULLID].intvalue;
options->perc_identity = (FloatLo) myargs[ARG_PERC_IDENT].floatvalue;
options->hsp_num_max = myargs[ARG_MAXHSP].intvalue;
if (!believe_query)
options->megablast_full_deflines = TRUE;
/*if (options->megablast_full_deflines)
believe_query = FALSE;*/
query_bsp_array = (BioseqPtr PNTR) MemNew((MAX_NUM_QUERIES+1)*sizeof(BioseqPtr));
sepp = (SeqEntryPtr PNTR) MemNew(MAX_NUM_QUERIES*sizeof(SeqEntryPtr));
StrCpy(prefix, "");
global_fp = outfp;
options->output = outfp;
if (traditional_formatting) {
if (align_view < 7) {
if (html) {
fprintf(outfp, "<HTML>\n<TITLE>MEGABLAST Search Results</TITLE>\n");
fprintf(outfp, "<BODY BGCOLOR=\"#FFFFFF\" LINK=\"#0000FF\" "
"VLINK=\"#660099\" ALINK=\"#660099\">\n");
fprintf(outfp, "<PRE>\n");
}
init_buff_ex(90);
BlastPrintVersionInfo("mgblast", html, outfp);
fprintf(outfp, "\n");
MegaBlastPrintReference(html, 90, outfp);
fprintf(outfp, "\n");
if(!PrintDbInformation(blast_database, !db_is_na, 70, outfp, html))
return 1;
free_buff();
#ifdef OS_UNIX
fprintf(global_fp, "%s", "Searching");
#endif
}
}
aip = NULL;
if (myargs[ARG_ASNOUT].strvalue != NULL) {
if ((aip = AsnIoOpen (myargs[ARG_ASNOUT].strvalue,"w")) == NULL) {
ErrPostEx(SEV_FATAL, 1, 0, "blast: Unable to open output file %s\n", myargs[ARG_ASNOUT].strvalue);
return 1;
}
}
else if (align_view == 10 || align_view == 11)
{
const char* mode = (align_view == 10) ? "w" : "wb";
if ((aip = AsnIoOpen (blast_outputfile, (char*) mode)) == NULL) {
ErrPostEx(SEV_FATAL, 1, 0, "blast: Unable to open output file %s\n", blast_outputfile);
return 1;
}
}
if (align_view == 7) {
xml_aip = AsnIoOpen(blast_outputfile, "wx");
}
if (myargs[ARG_QUERYLOC].strvalue) {
Int4 start, end;
Megablast_GetLoc(myargs[ARG_QUERYLOC].strvalue, &start, &end);
options->required_start = start - 1;
options->required_end = end -1;
}
done = FALSE;
while (!done) {
num_bsps = 0;
total_length = 0;
done = TRUE;
SeqMgrHoldIndexing(TRUE);
mask_slp = last_mask = NULL;
while ((sepp[num_bsps]=FastaToSeqEntryForDb(infp, query_is_na, NULL,
believe_query, prefix, &ctr,
&mask_slp)) != NULL) {
if (!lcase_masking) /* Lower case ignored */
mask_slp = SeqLocFree(mask_slp);
if (mask_slp) {
if (!last_mask)
options->query_lcase_mask = last_mask = mask_slp;
else {
last_mask->next = mask_slp;
last_mask = last_mask->next;
}
mask_slp = NULL;
}
query_bsp = NULL;
SeqEntryExplore(sepp[num_bsps], &query_bsp, FindNuc);
//debug:
/*
char query_buffer[255];
SeqIdWrite(query_bsp->id, query_buffer, PRINTID_FASTA_LONG, BUFFER_LENGTH);
fprintf(stderr, "===> query_buf=%s\n", query_buffer);
*/
if (query_bsp == NULL) {
ErrPostEx(SEV_FATAL, 1, 0, "Unable to obtain bioseq\n");
return 2;
}
source = BioSourceNew();
source->org = OrgRefNew();
source->org->orgname = OrgNameNew();
source->org->orgname->gcode = options->genetic_code;
ValNodeAddPointer(&(query_bsp->descr), Seq_descr_source, source);
query_bsp_array[num_bsps++] = query_bsp;
total_length += query_bsp->length;
if (total_length > myargs[ARG_MAXQUERY].intvalue ||
num_bsps >= MAX_NUM_QUERIES) {
done = FALSE;
break;
}
}
if (num_bsps == 0)
break;
SeqMgrHoldIndexing(FALSE);
other_returns = NULL;
error_returns = NULL;
if (out_type==MBLAST_ENDPOINTS)
seqalign_array = BioseqMegaBlastEngine(query_bsp_array, blast_program,
blast_database, options,
&other_returns, &error_returns,
dummy_callback, NULL, NULL, 0,
MegaBlastPrintEndpoints);
else if (out_type==MBLAST_SEGMENTS)
seqalign_array = BioseqMegaBlastEngine(query_bsp_array, blast_program,
blast_database, options,
&other_returns, &error_returns,
dummy_callback, NULL, NULL, 0,
MegaBlastPrintSegments);
else if (out_type==MBLAST_ALIGN_INFO) {
/* -- Geo mod: do not print header
PrintTabularOutputHeader(blast_database,
(num_bsps==1) ? query_bsp_array[0] : NULL,
NULL, "megablast", 0, believe_query,
global_fp);*/
seqalign_array = BioseqMegaBlastEngine(query_bsp_array, blast_program,
blast_database, options,
&other_returns, &error_returns,
dummy_callback, NULL, NULL, 0,
MegaBlastPrintAlignInfo);
} else if (out_type==MBLAST_ALIGNMENTS) {
seqalign_array = BioseqMegaBlastEngine(query_bsp_array, blast_program,
blast_database, options, &other_returns,
&error_returns, align_view < 7 ? tick_callback : NULL,
NULL, NULL, 0, NULL);
}
#ifdef OS_UNIX
fflush(global_fp);
#endif
if (error_returns) {
BlastErrorPrint(error_returns);
for (vnp = error_returns; vnp; vnp = vnp->next) {
BlastDestroyErrorMessage((BlastErrorMsgPtr)vnp->data.ptrvalue);
}
ValNodeFree(error_returns);
}
if (traditional_formatting) {
dbinfo = NULL;
ka_params = NULL;
ka_params_gap = NULL;
params_buffer = NULL;
mask_loc = NULL;
matrix = NULL;
for (vnp=other_returns; vnp; vnp = vnp->next) {
switch (vnp->choice) {
case TXDBINFO:
dbinfo = vnp->data.ptrvalue;
break;
case TXKABLK_NOGAP:
ka_params = vnp->data.ptrvalue;
break;
case TXKABLK_GAP:
ka_params_gap = vnp->data.ptrvalue;
break;
case TXPARAMETERS:
params_buffer = vnp->data.ptrvalue;
break;
case TXMATRIX:
matrix = vnp->data.ptrvalue;
break;
case SEQLOC_MASKING_NOTSET:
case SEQLOC_MASKING_PLUS1:
case SEQLOC_MASKING_PLUS2:
case SEQLOC_MASKING_PLUS3:
case SEQLOC_MASKING_MINUS1:
case SEQLOC_MASKING_MINUS2:
case SEQLOC_MASKING_MINUS3:
ValNodeAddPointer(&mask_loc, vnp->choice, vnp->data.ptrvalue);
break;
default:
break;
}
}
#ifdef OS_UNIX
if(align_view < 7) {
fprintf(global_fp, "%s\n", " done");
}
#endif
if (myargs[ARG_MASKEDQUERY].strvalue) {
if ((mqfp = FileOpen(myargs[ARG_MASKEDQUERY].strvalue, "w")) == NULL)
ErrPostEx(SEV_WARNING, 1, 0, "Unable to open file %s for masked query\n",
myargs[ARG_MASKEDQUERY].strvalue);
}
hits_found = FALSE;
mask_loc_start = next_mask_loc = mask_loc;
mask_loc = NULL;
if (align_view == 7) {
mbxp = PSIXmlInit(xml_aip, "megablast", blast_database,
options, query_bsp_array[0], 0);
}
if (seqalign_array) { //results returned back for processing
ReadDBBioseqFetchEnable ("megablast", blast_database, db_is_na, TRUE);
for (index=0; index<num_bsps; index++) {
seqalign = seqalign_array[index];
if (next_mask_loc &&
SeqIdComp(SeqLocId((SeqLocPtr)next_mask_loc->data.ptrvalue),
query_bsp_array[index]->id) == SIC_YES) {
mask_loc = (SeqLocPtr)
MemDup(next_mask_loc, sizeof(SeqLoc));
next_mask_loc = next_mask_loc->next;
mask_loc->next = NULL;
}
if (mqfp) {
/* convert mask locations from all sources into
a single seqloc */
mask_slp = NULL;
if (mask_loc)
mask_slp = blastMergeFilterLocs(mask_slp,
(SeqLocPtr)mask_loc->data.ptrvalue,
FALSE, 0, 0);
PrintMaskedSequence(query_bsp_array[index], mask_slp,
mqfp, 50, lcase_masking);
SeqLocSetFree(mask_slp);
}
if (seqalign==NULL) {
mask_loc = MemFree(mask_loc);
continue;
}
hits_found = TRUE;
if (align_view < 7) {
init_buff_ex(70);
AcknowledgeBlastQuery(query_bsp_array[index], 70, outfp,
believe_query, html);
free_buff();
}
if (align_view == 8 || align_view == 9) {
if (align_view == 9)
PrintTabularOutputHeader(blast_database,
query_bsp_array[index], NULL, blast_program, 0,
believe_query, global_fp);
/* debug:
char qbuf[512];
strcpy(qbuf, BioseqGetTitle(query_bsp_array[index]));
fprintf(stderr, "---> Here: query title=%s\n", qbuf);
*/
BlastPrintTabulatedResults(seqalign,
query_bsp_array[index], NULL, number_of_alignments,
blast_program, !options->gapped_calculation,
believe_query, 0, 0,
global_fp, (align_view == 9));
ObjMgrFreeCache(0);
SeqAlignSetFree(seqalign);
mask_loc = MemFree(mask_loc);
continue;
}
//Geo mod:
else if (align_view>=12) {
MGBlastPrintTab(seqalign,
query_bsp_array[index], number_of_alignments,
!options->gapped_calculation,
global_fp);
ObjMgrFreeCache(0);
SeqAlignSetFree(seqalign);
mask_loc = MemFree(mask_loc);
continue;
}
else if(align_view == 7) {
IterationPtr iterp;
iterp = BXMLBuildOneQueryIteration(seqalign,
NULL, FALSE,
!options->gapped_calculation, index,
NULL, query_bsp_array[index], mask_loc);
IterationAsnWrite(iterp, mbxp->aip, mbxp->atp);
AsnIoFlush(mbxp->aip);
IterationFree(iterp);
SeqAlignSetFree(seqalign);
mask_loc = MemFree(mask_loc);
continue;
}
seqannot = SeqAnnotNew();
seqannot->type = 2;
AddAlignInfoToSeqAnnot(seqannot, align_type);
seqannot->data = seqalign;
if (aip) {
SeqAnnotAsnWrite((SeqAnnotPtr) seqannot, aip, NULL);
AsnIoReset(aip);
}
if (outfp) { /* Uncacheing causes problems with ordinal nos. vs. gi's. */
prune = BlastPruneHitsFromSeqAlign(seqalign, number_of_descriptions, NULL);
ObjMgrSetHold();
init_buff_ex(85);
PrintDefLinesFromSeqAlign(prune->sap, 80,
outfp, print_options, FIRST_PASS, NULL);
free_buff();
prune = BlastPruneHitsFromSeqAlign(seqalign, number_of_alignments, prune);
seqannot->data = prune->sap;
if (align_view != 0)
ShowTextAlignFromAnnot(seqannot, 60, outfp, NULL,
NULL, align_options, NULL,
mask_loc, NULL);
else
ShowTextAlignFromAnnot(seqannot, 60, outfp, NULL, NULL, align_options, NULL, mask_loc, FormatScoreFunc);
seqannot->data = seqalign;
prune = BlastPruneSapStructDestruct(prune);
ObjMgrClearHold();
ObjMgrFreeCache(0);
}
seqannot = SeqAnnotFree(seqannot);
mask_loc = MemFree(mask_loc);
} /* End loop on seqaligns for different queries */
ReadDBBioseqFetchDisable();
}
if (mbxp != NULL) {
MBXmlClose(mbxp, other_returns, !options->gapped_calculation);
}
if (mqfp)
FileClose(mqfp);
if (!hits_found && align_view < 7)
fprintf(outfp, "\n\n ***** No hits found ******\n\n");
matrix = BLAST_MatrixDestruct(matrix);
if(html)
fprintf(outfp, "<PRE>\n");
init_buff_ex(85);
dbinfo_head = dbinfo;
if(align_view < 7) {
while (dbinfo) {
PrintDbReport(dbinfo, 70, outfp);
dbinfo = dbinfo->next;
}
}
dbinfo_head = TxDfDbInfoDestruct(dbinfo_head);
if (ka_params) {
if(align_view < 7)
PrintKAParameters(ka_params->Lambda, ka_params->K, ka_params->H, 70, outfp, FALSE);
MemFree(ka_params);
}
if (ka_params_gap) {
if(align_view < 7)
PrintKAParameters(ka_params_gap->Lambda, ka_params_gap->K, ka_params_gap->H, 70, outfp, TRUE);
MemFree(ka_params_gap);
}
if(align_view < 7)
PrintTildeSepLines(params_buffer, 70, outfp);
MemFree(params_buffer);
free_buff();
mask_loc = mask_loc_start;
while (mask_loc) {
SeqLocSetFree(mask_loc->data.ptrvalue);
mask_loc = mask_loc->next;
}
ValNodeFree(mask_loc_start);
} else { //not traditional formatting
/* Just destruct all other_returns parts */
for (vnp=other_returns; vnp; vnp = vnp->next) {
switch (vnp->choice) {
case TXDBINFO:
TxDfDbInfoDestruct(vnp->data.ptrvalue);
break;
case TXKABLK_NOGAP:
case TXKABLK_GAP:
case TXPARAMETERS:
MemFree(vnp->data.ptrvalue);
break;
case TXMATRIX:
BLAST_MatrixDestruct(vnp->data.ptrvalue);
break;
case SEQLOC_MASKING_NOTSET:
case SEQLOC_MASKING_PLUS1:
case SEQLOC_MASKING_PLUS2:
case SEQLOC_MASKING_PLUS3:
case SEQLOC_MASKING_MINUS1:
case SEQLOC_MASKING_MINUS2:
case SEQLOC_MASKING_MINUS3:
mask_loc = vnp->data.ptrvalue;
SeqLocSetFree(mask_loc);
default:
break;
}
}
}
other_returns = ValNodeFree(other_returns);
MemFree(seqalign_array);
options->query_lcase_mask =
SeqLocSetFree(options->query_lcase_mask);
/* Freeing SeqEntries can be very expensive, do this only if
this is not the last iteration of search */
if (!done) {
for (index=0; index<num_bsps; index++) {
sepp[index] = SeqEntryFree(sepp[index]);
query_bsp_array[index] = NULL;
}
}
total_processed += num_bsps;
} /* End of loop on complete searches */
aip = AsnIoClose(aip);
/*if (align_view == 7)
xml_aip = AsnIoClose(xml_aip);*/
if (align_view < 7 && html)
fprintf(outfp, "</PRE>\n</BODY>\n</HTML>\n");
if (align_view < 7 && myargs[ARG_LOGINFO].intvalue)
fprintf(outfp, "Mega BLAST run finished, processed %d queries\n",
total_processed);
MemFree(query_bsp_array);
MemFree(sepp);
MemFree(qgaps_buf);
MemFree(dbgaps_buf);
options = BLASTOptionDelete(options);
FileClose(infp);
FileClose(outfp);
return 0;
}
