Sequence LOGO

Since the formats from different dataset are different, it need differents script to be done.

Related files

• parse_signalp_euk.py
• parse_signalp_gram-.py
• parse_signalp_gram-.-.py
• parse_signalp_gram+.py
• parse_signalp_gram+.-.py
• parse_spds_euk.py
• parse_spds_gram-.py
• parse_spds_gram-.-.py
• parse_spds_gram+.py
• parse_spds_gram+.-.py

Explanation

• Prefix parse_ represents parsing the data
• signalp and spds stand for the two dataset signalP and SPDS17
• Suffix contains bio-categories, and if following .- represents positive sample and negative sample
  • Eg.
  • parse_signalp_euk.py: Parsing Eukaryotes data from the SignalP dataset
  • parse_signalp_gram+.py: Parsing Gram positive's positive samples data from the SignalP dataset
  • parse_signalp_gram+.-.py: Parsing Gram positive's negative samples data from the SignalP dataset

scripts

• fasta.py
  • to_fasta: Takes IDs and sequences, turn them in to fasta format
  • read_fasta: Takes file's path in fasta format, and read it into array
• fix_sequence.py
  • fix_sequence: Takes sequences and length, return fixed sequences which in the given length(In this work the length is 96)

Makefile

For convencienc, provide a Makefile

Usage: make TARGET TARGET: Start with the specific dataset, a ., the CATEGORY is coming after, if negative samples is expected, a .- is coming after eg. make spds.gram-: Parsing the positive samples of Gram-negative from the SPDS17 dataset

WebLogo

http://weblogo.threeplusone.com/ The website provides online and offline tools to produce sequence logo.

install

sudo easy_install weblogo

weblogo.py

The script automatically converts every data into sequenceLOGO

online service

http://weblogo.threeplusone.com/create.cgi, would be look like this

After

1. Select fasta file
2. Adjust expected scheme
3. Click to gernerate

supposed to get a sequence LOGO like this.