The use of multiple alphabets in kappa-gene immunoglobulin DNA sequence comparisons.

Karlin, Samuel; Ghandour, Ghassan

doi:10.1002/j.1460-2075.1985.tb03763.x

Cited by 16 publications

(7 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this presentation we shall be concerned with a related class of simple DNA---namely, with tracts composed of only two of the four bases. Two-base tracts can be grouped into three families: The purine.pyrimidine family (R.Y), The keto.imino family (K.M), and the strong and weak family (S;W; Karlin and Ghandour 1985). R.Y and K.M tracts are all A,G or G,T on one strand and T,C or A,C on the complementary one, while S;W tracts are either all A.T or all G.C.…”

Section: Introductionmentioning

confidence: 99%

The frequency of two-base tracts in eukaryotic genomes

Yagil

1993

J Mol Evol

View full text Add to dashboard Cite

The frequency of two-base tracts is surveyed in a wide range of eukaryotic genomes using the special program TRACTS. All three two-base families are surveyed: R.Y (A,G.C,T), K.M (A,C.G,T), and S;W (A.T and G.C). Data for the human beta-globin complex, for the tobacco chloroplast, and for 247 nt mammalian promoter regions are presented. All two-base tracts longer than three or four bases are overrepresented to an extent surpassing by far their occurrence in a randomized DNA population in the majority of the genomic regions analyzed; 20-30 long tracts are quite frequent, against the statistical odds. R.Y tracts are found at the largest excess, K.M tract to a slightly lesser extent, while S.W tracts are found at a moderate yet significant excess. The majority of the tracts manifest only a limited extent of tandem repeat structures. The idea that the two base tracts serve as unwinding elements is considered.

show abstract

Section: Introductionmentioning

confidence: 99%

The frequency of two-base tracts in eukaryotic genomes

Yagil

1993

J Mol Evol

View full text Add to dashboard Cite

show abstract

“…1). The R-Y alphabet comparisons reveal a striking abundance of substantial extensions of DNA block identities and new statistically significant block identities (14). All the statistically significant R-Y block identities are in close alignment relative to boundaries of the J gene segments or the C domain.…”

Section: The Use Of Multiple Alphabetsmentioning

confidence: 94%

Comparative statistics for DNA and protein sequences: multiple sequence analysis.

Karlin¹,

Ghandour²

1985

Proc. Natl. Acad. Sci. U.S.A.

View full text Add to dashboard Cite

Concepts and methods Proc. NaWl. Acad (i) Multiple Sequence Count Distribution of Common Word Occurrences. We define fk (vl, v2, ..., vs) = fk(v) = n, as the number of k-words that are represented across the sequences according to the vector v. We present the bivariate count distributions, along with permutation ranges, of common 8-words of the a-and f3-globin DNA sequences of human and mouse (Table 1). Similar results hold for the human and mouse P-globin and mouse a-and f3-globin comparisons (data not shown). The frequency of shared words is considerably higher than that obtained from corresponding permuted sequences. The number of words shared once between the human a-and p-globin sequences is also significantly high but closer to the permutation range. The fact that the human and mouse globin sequences decisively share k-words for all k 2 6 undoubtedly reflects the common function and evolution of these genes. The human a-and /3-globin sequences also show significant coincidence in k-word counts, but to a reduced degree, consistent with the precept of a more remote divergence time of the a-and 8-globin genes than the split times of the human and mouse lineages. The significant but reduced a-and f3-globin common word counts (k 2 6) perhaps also suggests partially differentiated structure or function between these globin genes.(ii) Multiple Sequence High Frequency (hf) Words. A criterion based on the cumulative number of occurrences across the s sequences prescribes a high frequency word of length k as a k-word satisfying 1t=1 vi 2 v*(k) or alternatively vi.I-*for i= 1, ..., s.hf shared k-words (k = 6-9) among Ig-K sequences ( Table 2). All the hf shared words of -7 base pairs (bp) relate to the "consensus" nonamer (GGTTTTT I lGT) postulated to play a control role in variable region (V)/joining region (J) rearrangement during B-cell differentiation (2-4). Among the three sequences, the mouse sequence tends to contain the most hf words of length -8 bp. Note that all hf 6-words of Table 2 can be formed by a single point mutation in a string of one-letter iterations with subsequent tandem duplications (cf. ref. 1). Interestingly, the common form of the polyadenylylation sequences (AATAAAj occurs nine times in each of the human and mouse sequences and six times in the rabbit sequence. This suggests that this oligonucleotide alone cannot be a decisive signal for transcription termination.(iii) Long Common Words. For s sequences we defined Krs as the length of the longest word occurring in at least r of the s sequences. With the aid ofTheorem I in ref. 1, the statistical significance of a long r-of-s shared word can be assessed. We present in Table 3 a number of examples of the several distinct longest r-of-s common words (block identities).(a) Virtually all the significant common oligonucleotides in the papova-, papilloma-, and the hepatitis viruses relate similar coding regions and are in alignment [i.e., spacings between successive significant block identities are about the same length for each sequence...

show abstract

“…DNA parameters and nucleotide proportions in each codon position were directly calculated from gene sequences. Calculations were also made for groupings of nucleotides into two-base categories: purines R (A and G) and pyrimidines Y (C and T), keto K (G and T) and amino M (A and C) nucleotides, and strong S (C and G) and weak W (A and T) bases (Karlin and Ghandour, 1985). The length indicates the number of nucleotides.…”

Section: Sequence-derived Variablesmentioning

confidence: 99%