BMC Bioinformatics

official impact factor 3.03

Open Access Research article

TransportTP: A two-phase classification approach for membrane transporter prediction and characterization

Haiquan Li, Vagner A Benedito, Michael K Udvardi and Patrick X Zhao*

Author Affiliations

Plant Biology Division, The Samuel Roberts Noble Foundation, Inc, Ardmore, OK 73401, USA

For all author emails, please log on.

BMC Bioinformatics 2009, 10:418 doi:10.1186/1471-2105-10-418

Published: 14 December 2009

Additional files

Additional file 1:

Comparative performance of TransportTP on non-model and model organisms. PDF displaying relative balanced accuracy, recall and precision of TransportTP on non-model organisms subtracted that on model organisms in leave-one-in cross-validations.

Format: PDF Size: 65KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data

Additional file 2:

Performance of TransportTP on transporter families covered by TransportDB. This PDF table displays recall, precision and balanced accuracy of TransportTP on transporter families covered by TransportDB, using the proteome of yeast for training, the proteomes of ten other organisms for testing, and 0.1 as e-value threshold.

Format: PDF Size: 84KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data

Additional file 3:

Complete validation results of TransportTP on seven training model-organisms. PDF displaying complete balanced accuracy, recall and precision of TransportTP in leave-on-in cross-validations at e-value threshold of 0.1, where one of the proteomes of the seven model organisms was used for training and the proteomes of ten other organisms were used for testing.

Format: PDF Size: 107KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data

Additional file 4:

Comparative results of TransportTP on non-model organisms using leave-multiple-in versus leave-one-in cross-validations. PDF displaying relative balanced accuracy, recall and precision of TransportTP on non-model organisms yielded by leave-multiple-in cross-validations subtracted that of leave-one-in cross-validations.

Format: PDF Size: 69KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data

Additional file 5:

Validation results of TransportTP on P. patens and C. reinhardtii using TransportDB as the benchmark database. This PDF displays validation results of TransportTP on P. patens and C. reinhardtii using arabidopsis for training and e-value thresholds between 10 and 0.00001 for homology search, and TransportDB for the benchmark database.

Format: PDF Size: 55KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data

Additional file 6:

Comparative performance of TransportTP with four alternative refining classifiers. This PDF displays the comparative recall versus precision, and balanced accuracy versus e-value thresholds among TransportTP with ensemble of balanced SVMs, balanced random forest, traditional SVM, and decision tree J48.

Format: PDF Size: 40KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data

Additional file 7:

Mapping of Pfam domains to Transporter Classification (TC) families/superfamilies. This PDF table displays the mapping of Pfam domains to Transporter Classification families or superfamilies, using 1) cross-links in Pfam database, 2) automatic mapping between Pfam domains and TC families/superfamilies and 3) manual curation.

Format: PDF Size: 151KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data

Additional file 8:

The keywords adopted in the text-mining program to detect non-transporter functions in Swiss-Prot database. This PDF displays the keywords adopted in the text-mining program to detect non-transporter functions hit by unknown proteins in Swiss-prot Database.

Format: PDF Size: 20KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data