MG7: A fast horizontally scalable tool based on cloud computing and graph databases for microbial community profiling.
2014
Methods: MG7 i s a n ope n s ource t ool i mplemented i n J ava a nd Scala, ba sed on c loud c omputing ( Amazon W eb S ervices). The g raph da ta platform B io4j ( www.bio4j.com) i s us ed f or r etrieving t axonomy r elated information, w hile N ispero ( http://ohnosequences.com/nispero) i s used f or distributing and coordinating compute tasks. Results: MG7 i s a n ope n-source, f ast a nd hor izontally s calable t ool f or community pr ofiling ba sed on t he a nalysis of 16S m etagenomics da ta. I t i s entirely c loud-based an d s pecifically d esigned t o t ake ad vantage o f i t: i t performs the community profiling of a sample starting from raw Illumina reads in a pproximately 1 ho ur, needing a pproximately t he s ame t ime for d oing t he same on hundreds of samples, adjusting automatically the computation capacity to the resources needed in each project. The taxonomic assignment can be done using a Best BLAST hit paradigm or a Lowest Common ancestor Paradigm; the user can choose between both assignment algorithms and setting the similarity parameters required for the assignment. As an output, MG7 generates the frequencies of all the identified taxa in any of the s amples i n t ab-separated value t ext f iles as well as i n t he s tandard B IOM format c ompliant w ith o ther m etagenomics to ols. T his o utput in cludes d irect assignment frequencies an d cu mulative f requencies b ased o n t he h ierarchical structure of t he t axonomy t ree. It a lso pr ovides w ith out put f iles s uitable f or generating heat-map representations. MG7 is an open-source tool available under the AGPLv3 license This project is funded in part by the ITN FP7 project INTERCROSSING (Grant 289974) a nd t he S panish C DTI ( Centro pa ra e l Desarrollo T ecnologico Industrial) grant NEXTMICRO, ref. IDI-20120242.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
0
References
0
Citations
NaN
KQI