script to get basic statistics from velvet contigs.fa

[mmmm]

could I get a script that can generate some basic statistics from the velvet output file (contigs.fa)

[sphil]

Please be more specific about what you need and what your understanding of "statistics from the velvet output" is. Do you mean N50, number of Contigs, longest, shortest....

[mmmm]

Thanks. to be more specific I would like to see the following statistics for velvet output fasta file:

-Statistics for contig lengths:
Min contig length
Max contig length
Mean contig length
standard deviation of contig length
Median contig length
N50 contig length

- Statistics for no. of contigs:
No. of contigs
No of contigs >=1kb
No. of contigs in N50

-Statistics for bases in the contigs:
No. of bases in all contigs
No. of bases in contigs >=1kb
GC content of contigs

-Simple Dinucleotide repeats:
No. of contigs with over 70% dinucleotide repeats
AT
CG
AC
TG
AG
TC

_Simple mononucleotide repeats:
No. of contigs with over 50% dinucleotide repeats
AA
TT
CC
GG

[GenoMax]

Take your pick:

https://github.com/ajmazurie/velvet-stats

http://milkweedgenome.org/?q=node/73

http://korflab.ucdavis.edu/datasets/Assemblathon/Assemblathon2/Basic_metrics/assemblathon_stats.pl

[sphil]

i don't which of your interest will be covered but have a look here velet std summary. I find the way they do it quite neat. However, i think you have to script something yourself to get all desired information.

[maasha]

Using Biopieces (www.biopieces.org):

Assembled contigs can be analyzed to get some stats using analyze_assembly. We include a filtering step to discard contigs shorter than 200 bases:

read_fasta -i contigs.fna |
grab -e "SEQ_LEN>=200" |
analyze_assembly -x

And the output:

N50: 9082
MAX: 52038
MIN: 200
MEAN: 4170
TOTAL: 3057214
COUNT: 733
---

There are many other options for stats as well. Look here:

https://code.google.com/p/biopieces/wiki/HowTo#Quality_assurance_and_cleaning_NGS_data