1 / 87

Methods for Repeat Detection In Nucleotide Sequences

Methods for Repeat Detection In Nucleotide Sequences. Gary Benson Computer Science, Biology, Bioinformatics Boston University gbenson@bu.edu. Outline. Classes of Repeats in DNA Tandem Repeats Techniques for finding repetitive sequence Tandem Repeats Database. Why Look at Repeats in DNA?.

miriam
Télécharger la présentation

Methods for Repeat Detection In Nucleotide Sequences

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Methods for Repeat Detection In Nucleotide Sequences Gary Benson Computer Science, Biology, Bioinformatics Boston University gbenson@bu.edu

  2. Outline • Classes of Repeats in DNA • Tandem Repeats • Techniques for finding repetitive sequence • Tandem Repeats Database

  3. Why Look at Repeats in DNA? • Repeats make up the largest portion of DNA. • coding sequence (~5% of human DNA) • repetitive sequence (>50% of human DNA)

  4. Classes of Repeats in DNA • Interspersed repeats: • Retrotransposons • Sines: • Lines: • LTRs • Transposons

  5. Classes of Repeats in DNA • Inverted repeats • Tandem repeats • Satellite repeats • microsatellites • minisatellites • VNTR (variable number of tandem repeats)

  6. Tandem Repeats A tandem repeat (TR) is any pattern of nucleotides that has been duplicated so that it appears several times in a row. For example, the sequence fragment below contains a tandem repeat of the trinucleotide cgt: tcgctggtcata cgt cgt cgt cgt cgt tacaaacgtcttccgt

  7. Approximate Tandem Repeats Shown are and a consensus pattern More typically, the tandem copies are only approximate due to mutations. Here is an alignment of copies from a human TR from Chromosome 5. 23.7 copies

  8. Why are tandem repeats interesting? • They are associated withhuman disease: Fragile-X mental retardation Myotonic dystrophy Huntington’s disease Friedreich’s ataxia Epilepsy Diabetes Ovarian cancer • They are often polymorphic, making them valuable genomic markers. • They are involved in gene regulationand often may contain transcription factor binding sites.

  9. LOCUS RATIGCA 4461 bp DNA ROD 18-APR-1994 DEFINITION Rat Ig germline epsilon H-chain gene C-region, 3' end. 2881 cgccccaagt aggcttcatc atgctctttg gtttagcaat agcccaaagc aagctatgca 2941 tccatctcag gcccagaggg atgaggagac cagaatcaag acatacccac gcccatccca 3001 cgcccaacca ccaaccacca gcacatcagg ttcacacacc tgagaccagt ggctcccatc 3061 acacacacac acacacacac acacacacac acacacacac acacacaagc ccgtacacat 3121 ccaccatatc cagagacaag tgtctgagtc tgagatacct ctgaggatca ccaatggcag 3181 agtcggccag cacctcagcc tccaggccaa tccttatact ttggcccact gcaggccatg 3241 agagatggag gaggtggagg cctgagctgt ggaaaaccag agacaggaag atggtctgta 3301 ctccaggcca atccttatac tttggcccac tgcaggccat gagagatgga ggaggtggag 3361 gcctgagctg tggaaaacca gagacaggaa gatggtctgt atggagagag tagtaaacca 3421 gattataggg agactgaggc aggagtagag ctcctacaag gccagtagtc taccttagag 3481 tcctataagt ctgggctggg agtccatgtg tcctgacttg ctcctcagat atcacaacca 3541 agattcctgg agccagagtg tgcatgcagg ccctagaaga aatgtggagc ttagagccct 3601 tcctggaggg ccctgggcac tctgaacaaa aggcaattct gtaggctgta tagaggcatc 3661 ctgtcagata cacacacaca tgcacacaca tacacacaca gagacacaga cacacacaca 3721 tgcccacaca catgcataca cacatgcaca cacatacaca cacagagaca cagacacaca 3781 catgcccaca cacatgcata cacacatgca tgcacacaca cacacacaca tacacataca 3841 cacacacaca cacaccccgc aggtagcctt catcatgctg tctagcgata gccctgctga 3901 gggtgggaga tactgggtca tggtgggcac cggagtagaa agagggaatg agcagtcagg 3961 gtcaggggaa aaggacatct gcctccaggg ctgaacagag acttggagca gtcccagagc 4021 aagtgggatg gggagctctg ccactccagt ttcaccagga ctgcctgaga ccagtgaggg

  10. LOCUS RATIGCA 4461 bp DNA ROD 18-APR-1994 DEFINITION Rat Ig germline epsilon H-chain gene C-region, 3' end. 2881 cgccccaagt aggcttcatc atgctctttg gtttagcaat agcccaaagc aagctatgca 2941 tccatctcag gcccagaggg atgaggagac cagaatcaag acatacccac gcccatccca 3001 cgcccaacca ccaaccacca gcacatcagg ttcacacacc tgagaccagt ggctcccatc 3061 acacacacac acacacacac acacacacac acacacacac acacacaagc ccgtacacat 3121 ccaccatatc cagagacaag tgtctgagtc tgagatacct ctgaggatca ccaatggcag 3181 agtcggccag cacctcagcc tccaggccaa tccttatact ttggcccact gcaggccatg 3241 agagatggag gaggtggagg cctgagctgt ggaaaaccag agacaggaag atggtctgta 3301 ctccaggcca atccttatac tttggcccac tgcaggccat gagagatgga ggaggtggag 3361 gcctgagctg tggaaaacca gagacaggaa gatggtctgt atggagagag tagtaaacca 3421 gattataggg agactgaggc aggagtagag ctcctacaag gccagtagtc taccttagag 3481 tcctataagt ctgggctggg agtccatgtg tcctgacttg ctcctcagat atcacaacca 3541 agattcctgg agccagagtg tgcatgcagg ccctagaaga aatgtggagc ttagagccct 3601 tcctggaggg ccctgggcac tctgaacaaa aggcaattct gtaggctgta tagaggcatc 3661 ctgtcagata cacacacaca tgcacacaca tacacacaca gagacacaga cacacacaca 3721 tgcccacaca catgcataca cacatgcaca cacatacaca cacagagaca cagacacaca 3781 catgcccaca cacatgcata cacacatgca tgcacacaca cacacacaca tacacataca 3841 cacacacaca cacaccccgc aggtagcctt catcatgctg tctagcgata gccctgctga 3901 gggtgggaga tactgggtca tggtgggcac cggagtagaa agagggaatg agcagtcagg 3961 gtcaggggaa aaggacatct gcctccaggg ctgaacagag acttggagca gtcccagagc 4021 aagtgggatg gggagctctg ccactccagt ttcaccagga ctgcctgaga ccagtgaggg

  11. Tandem Repeats Finder An online sequence analysis tool. OR A program to download and run locally. Data from TRF is listed as “simple repeats” at the UCSC genome browser website.

  12. Similarity Models Match/mismatch model – Sequences differ only by mismatches: AAAGCTTCGGAGTGCCCGA AATGCATCGGGGTGCCTGA Sequence 1: Sequence 2:

  13. Similarity Models Match/mismatch model – Sequences differ only by mismatches: AAAGCTTCGGAGTGCCCGA AATGCATCGGGGTGCCTGA 1101101111011111011 Alignments of similar sequences can be represented by bit strings (zeros and ones). Sequence 1: Sequence 2:

  14. Similarity Models Match/mismatch model – Sequences differ only by mismatches: One model parameter required: p = probability of matching letters in a column = probability of a 1 in the bit string Sometimes known as a Bernoulli (coin toss) model.

  15. Similarity Models Match/mismatch/indel model – adds indels to sequence differences: AAAGCTTCGG-AGT--GCCCGA AA-GCATCGGGAGTTAGCCTGA Sequence 1: Sequence 2:

  16. Similarity Models Match/mismatch/Indel model – adds indels to sequence differences: AAAGCTTCGG-AGT--GCCCGA AA-GCATCGGGAGTTAGCCTGA 1121101111211122111011 Alignments of sequences can be represented by strings of numbers in [0,1,2]. Sequence 1: Sequence 2:

  17. Similarity Models Match/mismatch/indel model – adds indels to sequence differences: At least two model parameters required: p = probability of matching letters in a column = probability of a 1 in the numerical string r = probability of an insertion or deletion = probability of a 2 or 3 in the numerical string

  18. Detecting Similar Sequences Methods for similarity detection involve some form of scanning the input sequences, usually, with a window of fixed size. Information about the contents of the window is stored. This is called indexing. 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 A C G T T G C A G T T G A C T G A C G

  19. Indexing The index is a list of all possible window contents together with a list, for each content, of where it occurs: AAA, AAC,….,ACG,…,CAG,…,CGT,…GAC,…,GTT,…,TGC,…,TTG,…TTT

  20. Building an Index First sequence: 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 A C G T T G C A G T T G A C T G A C G AAA, AAC,….,ACG,…,CAG,…,CGT,…GAC,…,GTT,…,TGC,…TTG,…TTT

  21. Building an Index 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 A C G T T G C A G T T G A C T G A C G AAA, AAC,….,ACG,…,CAG,…,CGT,…GAC,…,GTT,…,TGC,…TTG,…TTT 0

  22. Building an Index 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 A C G T T G C A G T T G A C T G A C G AAA, AAC,….,ACG,…,CAG,…,CGT,…GAC,…,GTT,…,TGC,…TTG,…TTT 0 1

  23. Building an Index 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 A C G T T G C A G T T G A C T G A C G AAA, AAC,….,ACG,…,CAG,…,CGT,…GAC,…,GTT,…,TGC,…TTG,…TTT 0 1 2

  24. Building an Index 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 A C G T T G C A G T T G A C T G A C G AAA, AAC,….,ACG,…,CAG,…,CGT,…GAC,…,GTT,…,TGC,…TTG,…TTT 0 1 2 3

  25. Building an Index 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 A C G T T G C A G T T G A C T G A C G AAA, AAC,….,ACG,…,CAG,…,CGT,…GAC,…,GTT,…,TGC,…TTG,…TTT 0 1 2 4 3

  26. Building an Index 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 A C G T T G C A G T T G A C T G A C G AAA, AAC,….,ACG,…,CAG,…,CGT,…GAC,…,GTT,…,TGC,…TTG,…TTT 0 1 2 4 3 8

  27. Building an Index 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 A C G T T G C A G T T G A C T G A C G AAA, AAC,….,ACG,…,CAG,…,CGT,…GAC,…,GTT,…,TGC,…TTG,…TTT 0 1 2 4 3 8 9

  28. Scanning a new sequence AAA, AAC,….,ACG,…,CAG,…,CGT,…GAC,…,GTT,…,TGC,…TTG,…TTT 0 1 2 4 3 8 9 T G C A G T T G . . . Second sequence:

  29. Scanning a new sequence AAA, AAC,….,ACG,…,CAG,…,CGT,…GAC,…,GTT,…,TGC,…TTG,…TTT 0 1 2 4 3 8 9 T G C A G T T G . . . Second sequence:

  30. Scanning a new sequence AAA, AAC,….,ACG,…,CAG,…,CGT,…GAC,…,GTT,…,TGC,…TTG,…TTT 0 1 2 4 3 8 9 T G C A G T T G . . . Second sequence:

  31. Scanning a new sequence AAA, AAC,….,ACG,…,CAG,…,CGT,…GAC,…,GTT,…,TGC,…TTG,…TTT 0 1 2 4 3 8 9 T G C A G T T G . . . Second sequence:

  32. Scanning a new sequence AAA, AAC,….,ACG,…,CAG,…,CGT,…GAC,…,GTT,…,TGC,…TTG,…TTT 0 1 2 4 3 8 9 T G C A G T T G . . . Second sequence:

  33. Test subject sequence at matching locations Indexed sequence: 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 A C G T T G C A G T T G A C T G A C G Subject seq: T G C A G T T G . . . Indexed sequence: 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 A C G T T G C A G T T G A C T G A C G Subject seq: T G C A G T T G . . .

  34. Interaction between the similarity model and the index Once a model is chosen and the index is built, two questions arise: • Is it possible to find a match using the window size chosen? • How many character matches are likely to be detected with the window size chosen?

  35. Q1: Is it possible to find a match? This is known as the waiting time problem. Waiting Time: How many consecutive positions must be examined until a run of k ones occurs.

  36. Q1: Is it possible to find a match? This is known as the waiting time problem. Waiting Time: How many consecutive positions must be examined until a run of k ones occurs. Specific sequence example: AAAGCTTCGGAGTGCCCGA AATGCATCGGGGTGCCTGA 1101101111011111011 Sequence 1: Sequence 2:

  37. Waiting Time Specific Example Sequence 1: AAAGCTTCGGAGTGCCCGA AATGCATCGGGGTGCCTGA 1101101111011111011 Sequence 2:

  38. Q1: Is it possible to find a match? Waiting Time: Given a Bernoulli sequence with generating probability p and length n, what is the probability that a run of k ones occurs? Randomly generated Bernoulli sequence using p: k 1110101111011011010 n

  39. Waiting Time Formulas These calculate the probability of a first occurrence of a run of k ones at every sequence length from 1 to n. for n ≥ 3, k = 3 F(111:n) = P(1)3 – F(111: n – 1) · P(1) – F(111: n – 2) · P(1)2 – ∑ k = 3 to n – 3 F(111: k) · P(1)3 where: F(111:n) is the probability of a first occurrence of 3 ones in a row at position n, P(1) is the model probability of a match.

  40. Waiting Time Formulas Predictions: If k = 3, p = .5, n = 12 • In what position [0..12] is it most likely to get a first occurrence of 3 ones in a row? • By what position will there be a cumulative probability of 30% to see a first occurrence of 3 ones in a row? • What is the likely cumulative probability of getting 3 ones in a row anywhere up to position 12?

  41. Coin Flip Experiment Goal: To estimate the probability of finding three heads in a row at least once when tossing a coin n times. The Experiment: Each student tosses a coin until the first occurrence of 3 heads in a row. This is one trial. Record the flip where the 3rd head is found. If twelve tosses do not produce three heads in a row, stop and begin a new trial. Repeat. EXAMPLE: Note that the first occurrence of three heads in a row is marked in bold. 1 2 3 4 5 6 7 8 9 10 11 12 Trial 1: H T H H T H H H Trial 2: T T H H H Trial 3: H T T H H T H H H Trial 3: H T T H H T H T H H T T

  42. Waiting Time Formulas Calculated probabilities:

  43. Q2: How many character matches will be detected? This is known as the coverage problem. Coverage: Given a Bernoulli sequence with generating probability p and length n, what is the probability distribution for number of ones contained in runs of k or more ones?

  44. Q2: How many character matches will be detected? Specific sequence example: Let k = 3, n = 19 AAAGCTTCGGAGTGCCCGA AATGCATCGGGGTGCCTGA 1101101111011111011 n Sequence 1: Sequence 2: k

  45. Q2: How many character matches will be detected? Specific sequence example: Let k = 3, n = 19 AAAGCTTCGGAGTGCCCGA AATGCATCGGGGTGCCTGA 1101101111011111011 n Sequence 1: Sequence 2: Total character matches detected is 9.

  46. Data Structure – modified Aho Corasick Tree Seed is 1*1**1. Tree represents all patterns obtained by replacing each * by either 0 or 1. Fail links in AC tree go to longest match between a string suffix and a prefix of a pattern.

  47. Recurrence Formula

  48. Q2: How many character matches will be detected?

  49. Finding Tandem Repeats:Basic Assumption We assume that two, mutated, adjacent copies of a pattern will contain runs of exact matches. d d T A T A C G T C T C C A C G G A

More Related