#MetaProb
##Abstract ###Motivation Sequencing technologies allow the sequencing of microbial communities directly from the environment without prior culturing. Taxonomic analysis of microbial communities, a process referred to as binning, is one of the most challenging task when analyzing metagenomic reads data. The major problems are the lack of taxonomically related genomes in existing reference databases, the uneven abundance ratio of species, and the limitations due to short read lengths and sequencing errors.
###Results MetaProb is a novel assembly-assisted tool for un-supervised metagenomic binning using spaced seed for graph construction. The novelty of MetaProb derives from solving a few important problems: how to divide reads into groups of independent reads, so that l-mer frequencies are not overestimated; how to convert l-mer counts into probabilistic sequence signatures, that will correct for variable distribution of l-mers, and for unbalanced groups of reads, in order to produce better estimates of the underlying genome statistic. We show that MetaProb is effective for both simulated and real datasets. It can accurately (with F-measures of 87 for short reads and 97 long reads) and efficiently bin short and long reads with varying abundance ratios.
#Installation
##Install dependencies Install in local directory the following library:
##Download Metaprob MetaProb v2
If you want to use spaced seeds download MetaProb-S:
MetaProb-S v2
##Compilation:
Open terminal and go to MetaProb/Release/ and then:
make all
###Option compilation
Y = Number of core
-jY
#Algorithm Option In the following paragraph is described the input file's structure and the parameters available in the MetaProb algorithm.
##File accepted and structure
File accepted have the following structure:
Structure file .fna example:
>IDENTIFICATION
ATAATTGGCAAGTGTTTTAGTCTTAGAGAGATTCTCTAAGTCTAACTTGAACTCAATTTGGAATCATTTCCCAATTTTTA
Structure .fastq exemple:
@IDENTIFICATION #0/1
CCCATGCCTTTAGCCAAATTCACGGTTTGATCACCCCTAAAACCAGCCAATATACCGAAGTGGAAGCCAGCATAAATGGCCTCAATATTACCGAAATGGAT
+
HBIIIIIIIHHDIHIGIIGGIHIIGIDIIIIBIHI@IIH@HIIHIIF5IIHEII>BDAHIBIEDBEIDG@HAEH*I@AEI=#CE?G17EEDHDEB@@?#8B
In this NEW VERSION, paired-end reads are passed to the algorithm in separeted file in which the reads are paired in the same order in which are writen. So we raccomend to control the paired-end read if they are paired in the correct manner.
##Parameter
-si File path single-end reads
-pi File paths paired-end reads
-dirOutput Path directory output files. Default: output/
-graphType 0 = pair read in Paired-End read with same ID, 1 = All read Single-End, 2 = All read Single-End and then union groups with Paired-End info. Default: 0
-numSp Number expected specie in file. Default: 2
-q Size of q-mer used to create graph adiacences: Default: 30
-m Threshold of shared q-mer to create graph adiacences. Default: 5
-ssize Max Seed size in each group. Default: 9000
-lmerFreq Size of L-mer used to compute feature vector. Default: 4
-feature Feature used to compute. Default: 1
-mg Only group mode activated (output only the groups created). Default: Not Active
-eK Estimate K for K-means algorithm. Default: Active unless you specify -numSp
Only for MetaProb-S:
-q Spaced qmer used for overlapping reads. Ex. 1***1*111. 1 is the simbol considered, any others are like *. Default: 111111111111111111111111111111
##Advanced Option
-iterMaxKmeans Max iteration of Kmeans algorithm. Default: 100
-timeMaxKmeans Max time in seconds of Kmeans algorithm. Default: 3600
#Feature available There are several features available to describe the information contained in the groups, but two of this are the best:
- For Short Read (bp < 300) -> -feature 1 = NORM_D2star_All_Read_Prob_Lmer_Euclidian
- For Long Read (bp > 300) -> -feature 2 = NORM_D2star_Group_Prob_Bernulli_Lmer_Euclidian
-feature 1 = NORM_D2star_All_Read_Prob_Lmer_Euclidian, //NORM_D2star_All_Read_Prob_Kmer + Euclidian norm
-feature 2 = NORM_D2star_Group_Prob_Bernulli_Lmer_Euclidian, //NORM_D2star_Group_Prob_Bernulli_Kmer + Euclidian norm
-feature 3 = NORM_BiMeta, //Bimeta's paper norm
-feature 4 = NORM_D2star_Group_Prob_Lmer, //probability of L-mer in a group, given seed read L-mer's count vector
-feature 5 = NORM_D2star_All_Read_Prob_Lmer, //probability of L-mer in collection, given seed read L-mer's count vectors
-feature 6 = NORM_D2star_Group_Prob_Bernulli_Lmer, //probability of L-mer with Bernulli model in a group
-feature 7 = NORM_D2star_All_Read_Prob_Bernoulli, //probability of L-mer with Bernulli model in collection
-feature 8 = NORM_D2star_All_Seed_Read_Prob_Bernoulli, //probability of L-mer with Bernulli model in collection seed read
-feature 9 = NORM_BiMeta_Euclidian, //NORM_Size_Seed + Euclidian norm
-feature 10 = NORM_D2star_Group_Prob_Lmer_Euclidian, //NORM_D2star_Group_Prob_Kmer + Euclidian norm
-feature 11 = NORM_D2star_All_Read_Prob_Bernoulli_Euclidian, //NORM_D2star_Prob_Bernoulli_All_Read + Euclidian norm
-feature 12 = NORM_D2star_All_Seed_Read_Prob_Bernoulli_Euclidian, //NORM_D2star_Prob_Bernoulli_All_Seed_Read + Euclidian norm
#Run
Calls algorithm where is compiled:
./MetaProb -si ../TestInputFile/long_example_1.fna -numSp 2 -feature 2 -m 45
./MetaProb -pi ../TestInputFile/short_example_1.fna.1 ../TestInputFile/short_example_1.fna.2 -numSp 2
./MetaProb -pi ../TestInputFile/short_example_2.fna.1 ../TestInputFile/short_example_2.fna.2 -numSp 2 -feature 1
##Best Parameter
- For Short Read (bp < 300) -> -feature 1 -m 5
- For Long Read (bp > 300) -> -feature 2 -m 45
#Test File
##Single-End Reads:
Single-End_dataset_part1
Single-End_dataset_part2
##Paired-End Reads split file
Paired-End_split_dataset_part01
Paired-End_split_dataset_part02
Paired-End_split_dataset_part03
Paired-End_split_dataset_part04
Paired-End_split_dataset_part05
Paired-End_split_dataset_part06
Paired-End_split_dataset_part07
Paired-End_split_dataset_part08
Paired-End_split_dataset_part09
Paired-End_split_dataset_part10
#Publication
S.Girotto, C.Pizzi, M.Comin
MetaProb: accurate metagenomic reads binning based on probabilistic sequence signatures
Bioinformatics 2016 32 (17): i567-i575
Special Issue European Conference on Computational Biology - ECCB 2016
DOI: https://doi.org/10.1093/bioinformatics/btw466
S.Girotto, M.Comin, C.Pizzi
Metagenomic reads binning with spaced seeds
Theoretical Computer Science 2017, Vol. 698, pp. 88-99
DOI: https://doi.org/10.1016/j.tcs.2017.05.023
#License MIT