Human–chimpanzee alignment: Ortholog exponentials and paralog power laws

Gao, Kun; Miller, Jonathan

doi:10.1016/j.compbiolchem.2014.08.010

Cited by 10 publications

(14 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since Figures 2E,F are in log-log scale, we have shown that repeats with mismatches have power-law distribution for both D and C . This power-law distribution for size is consistent with other studies: the self-alignment for smaller genomes shows similar power-law like distribution in Gao and Miller ( 2011 , 2014 ). We also draw power-law functions with the known exponents: 1/ D , 1/ D 2 , and 1/ D 3 for size distribution, and 1/C 3 for copy number distribution.…”

Section: Distribution Of Approximate Repeats In the Human Referencsupporting

confidence: 92%

Mappability and read length

Freudenberg

2014

Front. Genet.

View full text Add to dashboard Cite

Power-law distributions are the main functional form for the distribution of repeat size and repeat copy number in the human genome. When the genome is broken into fragments for sequencing, the limited size of fragments and reads may prevent an unique alignment of repeat sequences to the reference sequence. Repeats in the human genome can be as long as 104 bases, or 105 − 106 bases when allowing for mismatches between repeat units. Sequence reads from these regions are therefore unmappable when the read length is in the range of 103 bases. With a read length of 1000 bases, slightly more than 1% of the assembled genome, and slightly less than 1% of the 1 kb reads, are unmappable, excluding the unassembled portion of the human genome (8% in GRCh37/hg19). The slow decay (long tail) of the power-law function implies a diminishing return in converting unmappable regions/reads to become mappable with the increase of the read length, with the understanding that increasing read length will always move toward the direction of 100% mappability.

show abstract

Section: Distribution Of Approximate Repeats In the Human Referencsupporting

confidence: 92%

Mappability and read length

Freudenberg

2014

Front. Genet.

View full text Add to dashboard Cite

show abstract

“…Interestingly, in this asymptotic regime, the MLD exhibits a power-law tail M (r) ∼ r α (identified as a straight line in the double logarithmic plot), where the exponent α is close to −5 for exonic sequences. This is in contrast to the MLD of non-coding sequences, where the exponent α is close to −4 [9,10,16]. This property appears to be impressively reproducible in the comparison of various pairs of species (see Fig S1).…”

Section: Introductionmentioning

confidence: 79%

Comparing the Statistical Fate of Paralogous and Orthologous Sequences

Massip

Sheinman

Schbath³

et al. 2016

Preprint

View full text Add to dashboard Cite

Since several decades, sequence alignment is a widely used tool in bioinformatics. For instance, finding homologous sequences with known function in large databases is used to get insight into the function of non-annotated genomic regions. Very efficient tools, like BLAST have been developed to identify and rank possible homologous sequences. To estimate the significance of the homology, the ranking of alignment scores takes a background model for random sequences into account. Using this model one can estimate the probability to find two exactly matching subsequences by chance in two unrelated sequences. The corresponding probability for two homologous sequences is much higher allowing to identify them. Here we focus on the distribution of lengths of exact sequence matches in protein coding regions pairs of evolutionary distant genomes. We show that this distribution exhibits a power-law tail with exponent α = −5. Developing a simple model of sequence evolution by substitutions and segmental duplications, we show analytically that paralogous and orthologous gene pairs contribute differently to this distribution. Our model explains the differences observed in the comparison of coding and non-coding parts of genomes, thus providing with a better understanding of statistical properties of genomic sequences and their evolution.

show abstract

“…To obtain all non-embedded and embedded maximal matches between two genomes, we must first of all specify the matching criteria and compare their sequences either by intersection or by alignment. In this paper, our non-embedded and embedded maximal matches are obtained without loss of generality from 4-base exact matched genome intersections on pristine (un-repeatmasked) whole-genome sequences (see section 5.2 and supplementary material 1 for computation details); non-embedded and embedded maximal matches identified in a Lastz raw alignment are referred to as "non-nested and nested CMRs" in (Gao and Miller 2014). Alternatively, we also examine another set of maximal matches-maximal unique matches ("MUMs" for short) (Delcher et al 1999;Delcher et al 2002;Kurtz et al 2004)-as an approximation to the non-embedded maximal matches defined in this paper.…”

Section: Identifying Lineal Orthologs By Non-embedded Maximal Matchesmentioning

confidence: 99%

“…We compare our non-embedded maximal matches identified by intersection (nem for short, see section 5.2 and supplementary material 1 for computation details) with exact matches returned by Lastz net alignment (net for short, see (Gao and Miller 2011) and (Gao and Miller 2014) for computation details). Table 1 shows such comparisons for two pairs of genomes: human/chimp and human/mouse.…”

Section: Comparison On Nucleotide Level With a Lastz Net Alignmentmentioning

confidence: 99%

See 1 more Smart Citation

Primary orthologs from local sequence context

Gao

Miller

2020

BMC Bioinformatics

Self Cite

View full text Add to dashboard Cite

Context-dependent identification of orthologs customarily relies on conserved gene order or whole-genome sequence alignment. It is shown here that short-range context-as short as single maximal matches-also provides an effective means to identify orthologs within whole genomes. On pristine (un-repeatmasked) mammalian whole-genome assemblies we perform a genome "intersection" that in general consumes less than one thirtieth of the computation time required by commonly used methods for whole-genome alignment, and we extract "non-embedded maximal matches," maximal matches that are not embedded into other maximal matches, as potential orthologs. An ortholog identified via non-embedded maximal matches is analogous to a "positional ortholog" or a "primary ortholog" as defined in previous literature; such orthologs constitute homologs derived from the same direct ancestor whose ancestral positions in the genome are conserved. At the nucleotide level, non-embedded maximal matches recapitulate most exact matches identified by a Lastz net alignment. At the gene level, reciprocal best hits of genes containing non-embedded maximal matches recover one-to-one orthologs annotated by Ensembl Compara with high selectivity and high sensitivity; these reciprocal best hits additionally include putatively novel orthologs not found in Ensembl (e.g. over two thousand for human/chimpanzee). The method is especially suitable for genome-wide identification of orthologs.

show abstract

Human–chimpanzee alignment: Ortholog exponentials and paralog power laws

Cited by 10 publications

References 26 publications

Mappability and read length

Mappability and read length

Comparing the Statistical Fate of Paralogous and Orthologous Sequences

Primary orthologs from local sequence context

Contact Info

Product

Resources

About