int main( int argc, char *argv[] )
{
int i ;
int ret ;
Alignments alignments ;
Alignments clippedAlignments ;
Blocks blocks ;
Genome genome ;
char *genomeFile = NULL ;
if ( argc < 2 )
{
printf( "%s", usage ) ;
exit( 0 ) ;
}
minimumSupport = 2 ;
minimumEffectiveLength = 200 ;
kmerSize = 23 ;
breakN = 1 ;
minContigSize = 200 ;
prefix = NULL ;
VERBOSE = false ;
outputConnectionSequence = false ;
aggressiveMode = false ;
for ( i = 1 ; i < argc ; ++i )
{
if ( !strcmp( "-b", argv[i] ) )
{
alignments.Open( argv[i + 1]) ;
++i ;
}
else if ( !strcmp( "-o", argv[i] ) )
{
prefix = argv[i + 1] ;
++i ;
}
else if ( !strcmp( "-f", argv[i] ) )
{
genomeFile = argv[i + 1] ;
++i ;
}
else if ( !strcmp( "-ms", argv[i] ) )
{
minimumSupport = atoi( argv[i + 1] ) ;
++i ;
}
else if ( !strcmp( "-ml", argv[i] ) )
{
minimumEffectiveLength = atoi( argv[i + 1] ) ;
++i ;
}
else if ( !strcmp( "-k", argv[i] ) )
{
kmerSize = atoi( argv[i + 1] ) ;
++i ;
}
else if ( !strcmp( "-breakN", argv[i] ) )
{
breakN = atoi( argv[i + 1] ) ;
++i ;
}
else if ( !strcmp( "-minContigSize", argv[i] ) )
{
minContigSize = atoi( argv[i + 1] ) ;
++i ;
}
else if ( !strcmp( "-v", argv[i] ) )
{
VERBOSE = true ;
}
else if ( !strcmp( "-cs", argv[i] ) )
{
outputConnectionSequence = true ;
}
/*else if ( !strcmp( "-aggressive", argv[i] ) )
{
aggressiveMode = true ;
}*/
else if ( !strcmp( "-bc", argv[i] ) )
{
// So far, assume the input is from BWA mem
clippedAlignments.Open( argv[i + 1] ) ;
clippedAlignments.SetAllowSupplementary( true ) ;
++i ;
}
else
{
fprintf( stderr, "Unknown parameter: %s\n", argv[i] ) ;
exit( 1 ) ;
}
}
if ( !alignments.IsOpened() )
{
printf( "Must use -b to specify the bam file." ) ;
return 0 ;
}
if ( prefix != NULL )
{
char buffer[255] ;
sprintf( buffer, "%s.out", prefix ) ;
fpOut = fopen( buffer, "w" ) ;
}
else
{
char buffer[255] ;
prefix = strdup( "rascaf" ) ;
sprintf( buffer, "%s.out", prefix ) ;
fpOut = fopen( buffer, "w" ) ;
}
if ( genomeFile != NULL )
{
genome.Open( alignments, genomeFile ) ;
alignments.Rewind() ;
}
if ( outputConnectionSequence == true && genomeFile == NULL )
{
fprintf( stderr, "Must use -f to specify assembly file when using -cs\n" ) ;
exit( EXIT_FAILURE ) ;
}
// 74619
//printf( "%c\n", genome.GetNucleotide( 74619, 4 ) ) ;
//exit(0) ;
// Build the graph
ret = blocks.BuildExonBlocks( alignments, genome ) ;
alignments.Rewind() ;
fprintf( stderr, "Found %d exon blocks.\n", ret ) ;
if ( clippedAlignments.IsOpened() )
{
fprintf( stderr, "Extend exon blocks with clipped alignments.\n" ) ;
Blocks extendBlocks ;
extendBlocks.BuildExonBlocks( clippedAlignments, genome ) ;
clippedAlignments.Rewind() ;
ret = blocks.ExtendExonBlocks( extendBlocks ) ;
fprintf( stderr, "Found %d exon blocks after extension.\n", ret ) ;
}
blocks.GetAlignmentsInfo( alignments ) ;
alignments.Rewind() ;
ret = blocks.BuildGeneBlocks( alignments, genome ) ;
alignments.Rewind() ;
fprintf( stderr, "Found %d gene blocks.\n", ret ) ;
blocks.BuildGeneBlockGraph( alignments ) ;
if ( clippedAlignments.IsOpened() )
{
blocks.AddGeneBlockGraphByClippedAlignments( clippedAlignments ) ;
}
// Cleaning
blocks.CleanGeneBlockGraph( alignments, genome ) ;
// Scaffolding
Scaffold scaffold( blocks, genome ) ;
//scaffold.Init( blocks ) ;
int componentCnt = scaffold.BuildComponent() ;
fprintf( stderr, "Found %d non-trivial gene block components.\n", componentCnt ) ;
// Possible for parallelization
for ( i = 0 ; i < componentCnt ; ++i )
{
scaffold.ScaffoldComponent( i ) ;
}
scaffold.ScaffoldGenome() ;
// Output the command line
fprintf( fpOut, "command line: " ) ;
char *fullpath = (char *)malloc( sizeof( char ) * 4096 ) ;
for ( i = 0 ; i < argc ; ++i )
{
char c = ' ' ;
if ( i == argc - 1 )
c = '\n' ;
if ( i > 0 && !strcmp( argv[i - 1], "-b" ) )
{
if ( realpath( argv[i], fullpath ) == NULL )
{
fprintf( stderr, "Failed to resolve the path of file %s.\n", argv[i] ) ;
exit( 1 ) ;
}
fprintf( fpOut, "%s%c", fullpath, c ) ;
}
else if ( i > 0 && !strcmp( argv[i - 1], "-f" ) )
{
if ( realpath( argv[i], fullpath ) == NULL )
{
fprintf( stderr, "Failed to resolve the path of file %s.\n", argv[i] ) ;
exit( 1 ) ;
}
fprintf( fpOut, "%s%c", fullpath, c ) ;
}
else
fprintf( fpOut, "%s%c", argv[i], c ) ;
}
free( fullpath ) ;
scaffold.Output( fpOut, alignments ) ;
return 0 ;
}
