how to remove sequences with 0 and * from fast

[Fad2012]

Dear all

My fasta file contains thousands of sequences, and it looks like this:

>r1624.1
*
>r1624.2
*
>r1625.2
GAAGGAAATACTACCCAGTATCCTTCGTTGGCCCAACGTTCCAGTACATGCAGGCTAATAACTATTACCCAG
>r1625.1
AGCACGATTAAAGAGTAGCTACTTAAAAACCTAATCAAAATTATTTCATCTCTGGTGATTATCACTAGAAAT
>r1626.1
ATTTGCAACTACCACTTGGCCACTCAGGATGATTTGCAAAACGCAGTGAACGTCATGTGGAGTAGAGACCTC
>r1626.2
TCTGACATAAGGGACTTGTATGCTTACGAGGAAGAGGCCATGGAGCAGGGCATTTCAAACTATATTGAGTCA
>r1627.1
*
>r1627.2
*
>r1641.1
0
>r1641.2
0
>r1643.1
0
>r1643.2
0

how to remove those sequences with 0 and * ?

Many thanks

[Richard Finney]

cat file.fa | awk '{if ((p%2)==0) s=$0;else if (($1!="0")&&($1!="*")) printf "%s\n%s\n",s,$0; p++}'

[bioBob]

dont need cat for awk to work

awk 'BEGIN{i=0}{if(i==0){hdr=$0; i++}else{if(($0!="*")&&($0!="0"))print hdr"\n"$0; i=0}}' <in.fasta >out.fasta

[Richard Finney]

I win the "Useless use of cat award" ! ( http://partmaps.org/era/unix/award.html ). There's also a built-in line counter variable in awk which I didn't use so I guess I just had a bad idiom day.