// writeFlankSeqs writes fasta files containing the sequence of unmapped flanks
// identified in the primary hits provided. cutoff specifies the minimum sequence
// length to consider. left and right specify the filenames for the left and right
// flank fasta sequence files.
func writeFlankSeqs(reads string, hits hitSet, cutoff int, left, right string) error {
f, err := os.Open(reads)
if err != nil {
return err
}
defer f.Close()
lf, err := os.Create(left)
if err != nil {
return err
}
rf, err := os.Create(right)
if err != nil {
return err
}
r := fasta.NewReader(f, linear.NewSeq("", nil, alphabet.DNA))
sc := seqio.NewScanner(r)
for sc.Next() {
seq := sc.Seq().(*linear.Seq)
h, ok := hits[seq.Name()]
if !ok {
continue
}
all := seq.Seq
if h.qStart >= cutoff {
seq.Seq = all[:h.qStart]
_, err := fmt.Fprintf(lf, "%60a\n", seq)
if err != nil {
return err
}
}
if h.qLen-h.qEnd >= cutoff {
seq.Seq = all[h.qEnd:]
_, err := fmt.Fprintf(rf, "%60a\n", seq)
if err != nil {
return err
}
}
}
err = sc.Error()
if err != nil {
return err
}
err = lf.Close()
if err != nil {
return err
}
return rf.Close()
}
func main() {
flag.Parse()
if *in == "" || *typ < 0 || 2 < *typ {
flag.Usage()
os.Exit(1)
}
cfn := []func(s seq.Sequence, start, end int) (float64, error){
0: complexity.WF,
1: complexity.Entropic,
2: complexity.Z,
}[*typ]
f, err := os.Open(*in)
if err != nil {
log.Fatalf("failed to open %q: %v", *in, err)
}
defer f.Close()
sc := seqio.NewScanner(fasta.NewReader(f, linear.NewSeq("", nil, alphabet.DNAgapped)))
for sc.Next() {
seq := sc.Seq().(*linear.Seq)
// err is always nil for a linear.Seq Start() and End().
c, _ := cfn(seq, seq.Start(), seq.End())
if *dist {
fmt.Printf("%s\t%v\t%d\n", seq.Name(), c, seq.Len())
continue
}
if c >= *thresh {
fmt.Printf("%60a\n", seq)
}
}
if err := sc.Error(); err != nil {
log.Fatalf("error during fasta read: %v", err)
}
}