Frequently Asked Questions (and Answers)

1. How to use ZCURVE web server?

• Please visit ZCURVE at http://gepa.org.cn/zcurve. When you run the program, you only need to input your genome sequence in fasta format and choose some options. In most cases, you may remain the default options. After initiating the program, there is not any necessary human intervention and it could run automatically.

2. How to input the sequence and what is the output?

• For the input, we provide an example of E. coli genome, you just need to prepare a file contain sequence like that. The program could output four files if you click all the output options. One is the chromosomal coordinates of all predicted genes, one includes base sequences of all genes in fasta format, the third corresponds to amino acid sequences encoded by each predicted gene. The last output file contains essentiality of all predicted genes provided by our post-processing program Geptop. The larger of the essentiality score, the more important function the gene owns. The first three files could be outputted immediately, usually in one minute, whereas the essentiality information needs over one hour to output. So you need to leave your email address to the server and it will send the information to you automatically after the complete results are obtained.

3. What is the major difference of functions between ZCURVE 3.0 and the original version?

• As regards the function, the new version has the following improvements: (i) it has higher (about 2%) accuracy than the older version. (ii) it could predict essentiality scores for all predicted genes. (iii)it could be run online and not need to download and install on your standalone computer.

4. What is the accuracy of ZCURVE 3.0 for predicting coding potentials and function essentiality?

• It could achieve the accuracy of 93.7%, which is slightly higher than Glimmer 3.02, for predicting protein-coding genes in 422 rokaryotic genomes. It could obtain the AUC over 95% when scoring function essentiality of annotated genes in E. coli and B. subtilis, which are most thoroughly investigated model genomes and have the most accurate genome-scale essentiality data.

5. If I want to annotate one phage, plasmid or virus genome, could I still use ZCURVE 3.0?

• No. In fact, we have developed one program ZCURVE_V to perfrom those specific task and it is freely access on http://gepa.org.cn/zcurve_v.

• Please feel free to contact the corresponding author: Prof. Feng-Biao Guo (fbguo@uestc.edu.cn) for details.