I have amplified a functional gene with my environmental samples and already did the cloning and sequencing. I'm trying to cluster similar genes and I came across two term commonly used but I do not know the difference between the two as essentially both cluster similar genes into one consensus gene right? Can anybody explain to me the difference between the two?
A contig is a consensus from a "contiguous" alignment of reads, or the sequence resulting from the overlap of multiple smaller sequences...
So:
AAAGAGA
AGATATGCTA
CTAACCA
Creates the contig:
AAAGAGATATGCTAACCA
OTU means "Operational Taxonomic Unit" and is a vague term. In the context of metagenomic and studies using 16S (etc) amplicon sequencing it refers to a cluster of sequences which have similarity above some threshold (say 3% or 4% etc) and can be assumed to come from the same "taxonomic unit" and is treated as such. A taxonomic unit (here's the vague part) could be species, genus, order... Depending on the sequence variation within that group for the sequences captured. BLASTing the consensus of that cluster would then give you some information as to the identity of the OTU.
Following the format of the above example:
GAGAGTACT (centroid)
GAGATTACT
GAGAGTATT
Would all cluster together with a similarity threshold of 10%, as the bottom two sequences are within 10% similarity of the top sequence. In the case of USearch, I believe the "centroid" sequences (or the sequence chosen to based similarity off of for determining cluster membership) is based on frequency of that sequence in the data.
Read: http://www.drive5.com/usearch/manual/otu_definition.html from the creator of Usearch
Hopefully that helps.
Thanks Tyler for your explanation. So in my case. Say I have same functional genes to be clustered into consensus sequence using de novo assembly where the overlap sequences will be clustered together, the resulting contig should equal OTU, right?
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA