Quantitative synteny scoring improves homology inference and partitioning of gene families

Abstract

BackgroundClustering sequences into families has long been an important step in characterization of genes and proteins. There are many algorithms developed for this purpose, most of which are based on either direct similarity between gene pairs or some sort of network structure, where weights on edges of constructed graphs are based on similarity. However, conserved synteny is an important signal that can help distinguish homology and it has not been utilized to its fullest potential.ResultsHere, we present GenFamClust, a pipeline that combines the network properties of sequence similarity and synteny to assess homology relationship and merge known homologs into groups of gene families. GenFamClust identifies homologs in a more informed and accurate manner as compared to similarity based approaches. We tested our method against the Neighborhood Correlation method on two diverse datasets consisting of fully sequenced genomes of eukaryotes and synthetic data.ConclusionsThe results obtained from both datasets confirm that synteny helps determine homology and GenFamClust improves on Neighborhood Correlation method. The accuracy as well as the definition of synteny scores is the most valuable contribution of GenFamClust.

GenFamClust: an accurate, synteny-aware and reliable homology inference algorithm

“…The GFC method [ 36 ] takes two (possibly intersecting) sets of sequences as input: query data Q and reference data R . Q should consist of genes for which we want to infer homology and R is only used when computing similarity and synteny correlation scores.…”

“…GFC uses NC as its similarity measure and introduces a synteny correlation score , SyC , which in turn is inferred from a local synteny score , SyS as its synteny measure. We refer to the original work [ 36 ] for further details on parameter settings, method, comparison with NC, and other details.

Fig.

…”

“…We define synteny score S y S ( g 1 , g 2 ) between two genes g 1 and g 2 as where n ( g ) represents the set of neighbor genes, upstream or downstream of g , at most at distance k , on a chromosome or contig. In our previous study [ 36 ], we determined that k =5 is a suitable number of neighbouring genes upstream or downstream to consider for estimating local synteny between genes of Metazoa.…”

“…In our earlier work [ 36 ], we presented a pipeline that builds on NC and combines its sequence similarity graph structure with gene order information to assess homology relationships. We showed that this approach, called GenFamClust (GFC), was more accurate than NC [ 36 ], which in turn has been shown to be better than BLAST scores [ 25 ].…”

“…We performed a rudimentary accuracy check for homologs inferred from GFC and NC in our previous study [ 36 ] and NC has been compared with BLAST for accuracy [ 25 ]. Gene families inferred by applying clustering algorithms on homologs inferred from BLAST, NC or GFC have not been checked for other cluster properties.…”