FAQ

Microbiome Search Engine is a search-able database of microbiome for in-depth data mining of microbial community studies by data integration and data selection. It contains large scale of microbial community samples with sequences, meta-data and analysis results. Based on the query sample(s), Microbiome Search Engine will provide the matched entry sample(s) in the database with very similar community structures in an ultra-fast speed. In addition, samples in the database can also be sorted, selected, exported and downloaded by their sampling meta-data information (meta-data).

This reference microbiome database is being constantly updated and will be released periodically.



Microbiome Search Engine accepts the query sample(s) that profiled by Parallel-META 3 for sequences. The latest Parallel-META 3 is available at http://bioinfo.single-cell.cn/parallel-meta.html

Sequence should be in FASTA/FASTQ format (eg, sample1.fa), then the command line for OTU profiling:

$PM-parallel-meta -r sample1.fa -o sample1.out -f F# for 16S rRNA sequences

or

$PM-parallel-meta -m sample1.fa -o sample1.out -f F# for shotgun sequences

Then in the output directory "sample1.out", the parsed taxonomy result file "classification.txt" is qualified for upload and search.This file contains two columns: the OTUs (greengene 13-8 on 97% level) and their sequence numbers.

Example:

#Database_OTU
Count
38922072
82487620
4414476830


Microbiome Search Engine accepts the samples in Parallel-META 31 results format. Sequences of samples should be analysed by Parallel-META 3. (See "How to pre-process your query sample")

In "Query" secion, click the "Select" button choose the file you want to search, and then pressen press "Upload" button to submit. There are 2 advanced options:

  • Maximum matched sample number: The max matched sample number for visualization and download, default is 10;
  • Minimum similarity:The min similarity value between input query and matched sample(s), default is 0.85. The similarity varies from 0.0-1.0, and higher similarity indicates more community structure shared by 2 samples.


The similarity between the query sample and the matched samples is calculated by Meta-Storms2, which incorporated both the relative abundance of each individual organism and the phylogeny architecture among organisms in the community on OTU level. The similarity varies from 0.0-1.0, and higher similarity indicates more community structure shared by the samples. Then the shared community structure of your input query sample and the matched samples (s) can be illustrated in a bar chart on phylum level and displayed in the result page. In addition, the searching results can also be downloaded with full community information including OTU table and Multi-Sample-View1, which could be further analysed in Parallel-META 31.

Microbiome Novelty Score (MNS) is proposed to evaluate the compositional uniqueness of a microbiome sample when compared to all microbiomes in the database. A higher MNS means lower similarity to those microbiomes that have previously been sampled, suggesting higher novelty. If a microbiome has MNS ≥ 0.12 (mean MNS of 2010) then it is considered as "Novel" sample. MNS starts from 2010, and is not applicable for samples produced/published before 2010. Click here to check all Novel samples.

Microbiome Attention Score (MAS) is proposed to measures the connectivity of a given sample to all subsequent samples in the database. A higher MAS means more samples with similarity or samples with higher similarity had been sequenced, suggesting higher attention from the scientific community for this input sample. If a microbiome has MAS ≥ 10 (top 20%) then it is considered as "High-attention" sample. MAS starts from 2010, and is not applicable for samples produced/published before 2010. Click here to check all High-attention samples.

If a microbiome has the two attributes of a) MNS ≥ 0.12 when first sequenced and b) MAS ≥ 10, then it is considered as “Focus” samples. Click here to check all Focus samples. The combined MNS and MAS are quantitatively measured by Microbiome Focus Index (MFI):

MFI = MNS * MAS


Microbiome Novelty Score (MNS) is proposed to evaluate the compositional uniqueness of a microbiome sample when compared to all microbiomes in the database. A higher MNS means lower similarity to those microbiomes that have previously been sampled, suggesting higher novelty. If a microbiome has MNS ≥ 0.12 (mean MNS of 2010) then it is considered as "Novel" sample. MNS starts from 2010, and is not applicable for samples produced/published before 2010. Click here to check all Novel samples.

Microbiome Attention Score (MAS) is proposed to measures the connectivity of a given sample to all subsequent samples in the database. A higher MAS means more samples with similarity or samples with higher similarity had been sequenced, suggesting higher attention from the scientific community for this input sample. If a microbiome has MAS ≥ 10 (top 20%) then it is considered as "High-attention" sample. MAS starts from 2010, and is not applicable for samples produced/published before 2010. Click here to check all High-attention samples.

If a microbiome has the two attributes of a) MNS ≥ 0.12 when first sequenced and b) MAS ≥ 10, then it is considered as “Focus” samples. Click here to check all Focus samples. The combined MNS and MAS are quantitatively measured by Microbiome Focus Index (MFI):

MFI = MNS * MAS


The Microbiome Search Engine online system also provides the complete database scores and meta-data, the search engine kernel standalone software (Meta-Storms 2) and the sequence profiling software (Parallel-META 3) in Download.





Help Desk

Email*:

Question*: