See the tool's GitHub repository, for the most updated information.
LocTree3 is an enhanced version of LocTree2
Prediction of protein subcellular localization is an important step towards elucidating protein function. LocTree2 is our previously published de-novo prediction method that classifies eukaryotic proteins in 18 localization classes, bacterial in 6 and archaeal in 3 classes. Since its development, LocTree2 has performed on a par with or better than any other state-of-the-art method. LocTree3 is a publicly available web server for LocTree2 and its extended version. This extension allows for homology inference if close homologs are available. Evaluated on a redundancy reduced set of 1682 eukaryotic proteins, the new method LocTree3 outperformed its predecessor reaching the overall accuracy Q18 = 80 ± 4%. On a set of sequence unique 479 bacterial proteins the overall prediction accuracy of LocTree3 was Q6 = 89 ± 4%.
The new web server provides:
- a fast resource for localization prediction in all domains of life
- free access for all users without login requirement
- informative visualization of our predictions
- prediction confidence for each result
- an additional homology inference step
- alignments and crosslinks for close homologs
- localization predictions for over 1000 completely sequenced organisms
LocTree3 is an extension of LocTree2 that is a hierarchical system of Support Vector Machines (SVM) inspired by the sorting machinery in the cell. The predictions with SVMs are made through searches of k-consecutive residues in proteins of experimental localization annotations. The improved version LocTree3 adds a module for inferring localization information from experimentally annottaed sequence homologs using PSI-BLAST. In the absence of significant PSI-BLAST hits, LocTree2 is used.
The input to the server is:
1. one or more fasta-formatted protein sequences. The sequences must be in one-letter amino acid code (not case-sensitive). The allowed amino acids are: ACDEFGHIKLMNPQRSTVWY and X (unknown). Example.
2. the domain of life: because LocTree3 predicts different localization classes for proteins from different domains of life, the domain must be chosen correctly (default: Eukarya)
3. e-mail address: a notification for completed prediction result and the access link are sent to the emai address (Optional)
For every query protein, result contains four basic values:
1. the protein identifier as provided by the user
2. the reliability score of a prediction on a 0-100 scale with 100 being the most confident prediction
3. single predicted localization class
4. GO term(s) and GO identifier(s) matching the predicted class.
Every result is supported by the information on whether it comes from of a PSI-BLAST homology search or a LocTree2 de novo prediction. In case of former, the web site provides ‘per click’ on the prediction result the experimental evidence (i.e. SWISS-PROT annotation) of the best hit and its PSI-BLAST alignment to the query protein. In case of latter, ‘the click’ on the result will forward to the visual representation of the prediction, providing decision tree with values at each of the decision points leading to the final reliability score. In addition, every result is supported by a schematic representation of the biological cell highlighting the predicted localization.
Every prediction result is supported by a Reliability Index (RI) measuring the strength of a prediction. The RI is a value between 0 and 100, with 100 denoting the most confident predictions.
We rigorously evaluated the reliability of LocTree3 predictions on a non-redundant test set of proteins (Fig. 1). We observed that 50% of proteins with the highest reliability were predicted for bacteria at RI>80 at an overall accuracy Q6=95% (Fig. 2; gray arrow) and for eukaryotes at RI>65 at Q18=95% (Fig. 2; black arrow).
- Q6 is six-state accuracy for predicting localization to six classes
- Q18 is eighteen-state accuracy
LocTree3 is built to run a homology-based PSI-BLAST; if no hit is identified then a de-novo LocTree2 prediction is used for localization annotation.
While PSI-BLAST searches are fast, LocTree2 runtime depends on the protein domain and the number of query protein sequences. We measured LocTree2 runtime on a Dell M605 machine with a Six-Core AMD Opteron processor (2.4 GHz, 6MB and 75W ACP) running on Linux.
|1 Sequence||100 Sequences||500 Sequences||1000 Sequences||3000 Sequences||5000 Sequences||10000 Sequences|
Note: to increase server's response time we store all PSI-BLAST profile files (required for LocTree2) in the PredictProtein cache (current size: results for >11Mio sequences). These can be retrieved from the cache very fast. For novel protein sequences for which we don't have PSI-BLAST profiles in the cache the runtimes increases substantially.
- The LocTree3 web server is available at https://rostlab.org/services/loctree3/
- LocTree3 predictions can also be accessed through the PredictProtein service
- Standalone version of LocTree3 can be downloaded as a Debian package here
- LocTree3 is available on GitHub here
- And also in the Elixir registry
Data sets used for development and evaluation of LocTree3 can be accessed here.
LocTree3 prediction of localization
Goldberg T, Hecht M, Hamp T, Karl T, Yachdav G, Ahmed N, Altermann U, Angerer P, Ansorge S, Balasz K, Bernhofer M, Betz A, Cizmadija L, Do KT, Gerke J, Greil R, Joerdens V, Hastreiter M, Hembach K, Herzog M, Kalemanov M, Kluge M, Meier A, Nasir H, Neumaier U, Prade V, Reeb J, Sorokoumov A, Troshani I, Vorberg S, Waldraff S, Zierer J, Nielsen H, Rost B.
LocTree2 predicts localization for all domains of life
Goldberg T, Hamp T, Rost B.
For questions, please contact email@example.com