Fequently Asked Questions

What do I need to run a MetaProx search?

To run the Search Tool you will require Java 8 or greater to be installed on your machine and we recommend installing the latest version of the Java Runtime Environment (JRE) to ensure optimal performance. You will also need a display with a resolution of at least 800 X 650 pixels. If you are running an online search then you will additionally require at least 400MB of available memory. Online searches invlove the use of Java Web Start technology which will require a browser that is capable of launching this type of application, such as Google Chrome, Internet Explorer, or Mozilla Firefox. Note, using a Windows High Contrast Color Scheme might cause the MetaProx website to be rendered incorrectly thereby causing an unusual and/or illegible appearance.

Why are there two search modes, by SOURCE and by LENGTH?

By default, proximons are sorted by descending length (i.e. the number of member genes, not sheer length in base pairs) because past research has demonstrated that increasing length is associated with greater confidence that a prediction represents a true operon. If specific metagenomes have been selected for a search then the search will proceed using indexing by SOURCE that will find the longest qualifying proximons from the selection. If the results queue becomes full, the search will continue in order to replace current hits in the queue with potential hits from longer proximons, thereby returning the longest qualifying hits from the selected metagenomes. If no specific metagenomes have been selected for a search then the search will include all available data and proceed using indexing by LENGTH. If the results queue becomes full, this allows the search to truncate its scheduled traversal because it is guaranteed that the longest qualifying proximons have already been found.

Why did I get a "java.lang.OutOfMemoryError" error?

All activities executed by the Search Tool occur by way of the Java Virtual Machine (JVM). As a result, JVM state directly governs Search Tool performance. When the JVM cannot allocate an object because it is out of memory and no more memory can be made available by the garbage collector, a memory error will be thrown, likely halting the activity of the Search Tool. To avoid this, the Search Tool has been designed to monitor free memory and freeze the size of its results queue if free memory drops below 30%. However, a variety of conditions can sometimes cause spiking drops in free memory that can result in sluggish performance or memory errors. If this occurs, relaunch the Search Tool using an increased heap size.

Why did I get a "Could not find requested resource" error?

This error occurs when the Search Tool is unable to retrieve supporting data that is not directly involved a database search. This error can occur as a result of either of the two following situations:

  • Connectivity: If your network connection is interrupted or lost, especially during application startup, then required data will not be obtained by the Search Tool. If this occurs, restore your connection and relaunch the application.
  • Compatibility: Periodically updated versions of either the MetaProx database or the Search Tool will be released. Depending on the associated changes, it is possible that the Search Tool might not recognize portions of the database. If this occurs, Download the latest version of the Search Tool.

Why did I get a "Could not connect to requested block" error?

This error occurs when the Search Tool is unable to search a specific portion of the database. This error can occur as a result of either of the two following situations:

  • Connectivity: If your network connection is interrupted or lost, especially during a search, then the MetaProx database will not be available to the Search Tool. If this occurs, restore your connection and relaunch the application.
  • Compatibility: Periodically updated versions of either the MetaProx database or the Search Tool will be released. Depending on the associated changes, it is possible that the Search Tool might not recognize portions of the database. If this occurs, Download the latest version of the Search Tool.

Can I get my own copy of the database?

Not yet. The current version of the Search Tool has been hard coded to connect exclusively to the online MetaProx database. However, we realize that certain research pipelines might be improved by the option to connect to a local database and therefore we plan to release a downloadable version of the database in the near future. Note, this will be in conjunction with the release of a new version of the Search Tool that will allow selection of the data connection.

Can I get source code?

No. The secure maintenance of the MetaProx database prevents the release of full source code. However, in order to support the use and distribution of serialized data, we are planning to release an Application Programming Interface (API) for the objects used by MetaProx. The API will allow users to easily restore saved data and invoke a variety of methods intended to programmatically extract information from proximons and their respective member genes.

Where does the data used for proximon predictions come from?

The proximon data and corresponding metagenomic genes that are currently included in MetaProx were derived from data obtained from the Integrated Microbial Genomes with Microbiome Samples (IMG/M) metagenomics database. This data contains the features of metagenomic genes obtained from various environments and locations, including their coordinates on their respective scaffold, as well as a variety of functional annotations.

How reliable are the predicted proximons?

Proximons were predicted from available metagenomic gene coordinates using a previously published method for identifying metagenomic operon candidates based on the intergenic distances between adjacent co-directional genes. All proximons included in MetaProx were obtained using a minimum threshold of confidence that is equivalent to a positive predictive value of 0.90. Therefore, based on evidence from known operons of Escherichia coli K12, 90% of the predicted proximons are expected to represent true metagenomic operons.

How do I cite MetaProx?

Please cite the following publication: Vey G & Charles TC (2014) MetaProx: the database of metagenomic proximons. Database (Oxford) 2014: bau097.