The number of expected nsSNVs is calculated using the following equation: S expected = S input L input L query where L input S input and L query depend on the type of the analysis.
(ii) L query and L subject distances are less than 500 nt and more than -7 nt (cases of partial overlaps).
L input is the sum of sizes of all the proteins in this protein set and L query is the size of the entire proteome.
L input is the sum of sizes of these sites found in various proteins and L query is the sum of sizes of these proteins that contain these sites.
L input is the sum of sizes of that domain found in various proteins and L query is the sum of sizes of these proteins that contain the domain.
Notably, Blast (http://blast.ncbi.nlm.nih.gov/Blast.cgi, last accessed October 28, 2013) and Blat (http://genome.ucsc.edu/cgi-bin/hgBlat command=start, last accessed October 28, 2013) rat genome searches did not reveal a perfect match for the L-Snord115 query sequence as well as its shorter form.ucsc.edu/cgi-bin/hgBlat command=start
'SQL' is usually pronounced 'S-Q-L' (Structured Query Language).
Thus, for a query L of 100 bp, M = 13.5∀7.2 for a single copy gene, where the Poisson distribution predicts the range based upon a 95% confidence interval.
In our implementation, for fast evaluation of the similarity functions, we compute the profiles, consisting of f Q k and f Q k, Q l for all query position k < l, prior to the actual scanning.
In the miRanda algorithm, the n-th valid alignment K q k, r l n between query q k and r l is represented by a tetrad K q k, r l n = { D i *, j *, n, i *, j * }, where i*, j* are the end positions of query and reference, respectively, and D i,j are the corresponding alignment scores.
The computational cost for approximate matching text comparison can be estimated by: (2) where c1 is some constant, N is the average number of records for a query, L is the average word sequence length (in journal name, title or author list) and W is the average word length.
