1. Title: Protein Localization Sites


2. Creator and Maintainer:
	     Kenta Nakai
             Institue of Molecular and Cellular Biology
	     Osaka, University
	     1-3 Yamada-oka, Suita 565 Japan
	     nakai@imcb.osaka-u.ac.jp
             http://www.imcb.osaka-u.ac.jp/nakai/psort.html
   Donor: Paul Horton (paulh@cs.berkeley.edu)
   Date:  September, 1996


3. Past Usage.
Reference: "A Probablistic Classification System for Predicting the Cellular 
           Localization Sites of Proteins", Paul Horton & Kenta Nakai,
           Intelligent Systems in Molecular Biology, 109-115.
	   St. Louis, USA 1996.
Results: 55% for Yeast data, %81 for E.coli with an ad hoc structured
	 probability model. Also similar accuracy for Binary Decision Tree and
	 Bayesian Classifier methods applied by the same authors in
	 unpublished results.

Predicted Attribute: Localization site of protein. ( non-numeric ).


4. The references below describe a predecessor to this dataset and its 
development. They also give results (not cross-validated) for classification 
by a rule-based expert system with that version of the dataset.

Reference: "Expert Sytem for Predicting Protein Localization Sites in 
           Gram-Negative Bacteria", Kenta Nakai & Minoru Kanehisa,  
           PROTEINS: Structure, Function, and Genetics 11:95-110, 1991.

Reference: "A Knowledge Base for Predicting Protein Localization Sites in
	   Eukaryotic Cells", Kenta Nakai & Minoru Kanehisa, 
	   Genomics 14:897-911, 1992.


5. Number of Instances:  1484 for the Yeast dataset.

6. Number of Attributes.
         for Yeast dataset:   9 ( 8 predictive, 1 name )
	     
7. Attribute Information.
  1.  Sequence Name: Accession number for the SWISS-PROT database
  2.  mcg: McGeoch's method for signal sequence recognition.
  3.  gvh: von Heijne's method for signal sequence recognition.
  4.  alm: Score of the ALOM membrane spanning region prediction program.
  5.  mit: Score of discriminant analysis of the amino acid content of
	   the N-terminal region (20 residues long) of mitochondrial and 
           non-mitochondrial proteins.
  6.  erl: Presence of "HDEL" substring (thought to act as a signal for
	   retention in the endoplasmic reticulum lumen). Binary attribute.
  7.  pox: Peroxisomal targeting signal in the C-terminus.
  8.  vac: Score of discriminant analysis of the amino acid content of
           vacuolar and extracellular proteins.
  9.  nuc: Score of discriminant analysis of nuclear localization signals
	   of nuclear and non-nuclear proteins.


8. Missing Attribute Values: None.


9. Class Distribution. The class is the localization site. Please see Nakai &
		       Kanehisa referenced above for more details.
  CYT (cytosolic or cytoskeletal)                    463
  NUC (nuclear)                                      429
  MIT (mitochondrial)                                244
  ME3 (membrane protein, no N-terminal signal)       163
  ME2 (membrane protein, uncleaved signal)            51
  ME1 (membrane protein, cleaved signal)              44
  EXC (extracellular)                                 37
  VAC (vacuolar)                                      30
  POX (peroxisomal)                                   20
  ERL (endoplasmic reticulum lumen)                    5