CONTENTS Introduction Analysis of individual sequences Secondary structure prediction Pairwise sequence comparison Database searching I: single heuristic algorithms Alignment and search statistics Multiple sequence alignment Multiple alignment and database searching Protein families and protein domains Conclusion
INTRODUCTION DNA STORES AND PASSES ON GENETIC INFORMATION FROM ONE GENERATION TO ANOTHER. THE STEPS OF THE LADDER ARE MADE OF PAIRS OF NITROGEN BASES: 1. ADENOSINE = A 2. GUANOSINE = G 3. CYTIDINE = C 4. THYMIDINE = T
INTRODUCTION DNA HELD BY HYDROGEN BONDS
INTRODUCTION STOP 2 START 2 STOP 1 START 1
INTRODUCTION RIBOSOME KNOWS TO WHICH AMINO ACID A CODON CORRESPONDS
INTRODUCTION PRIMARY STRUCTURE OF A PROTEIN = A CHAIN OF AMINO ACIDS
INTRODUCTION SECONDARY STRUCTURE -> 3RD DIMENSION
ANALYSIS OF INDIVIDUAL SEQUENCES AMINO ACIDS 1. NON-POLAR AND NEUTRAL 2. POLAR AND NEUTRAL 3. ACID AND POLAR 4. BASIC AND POLAR
ANALYSIS OF INDIVIDUAL SEQUENCES HYDROPHOBICITY - = HYDROPHILIC
ANALYSIS OF INDIVIDUAL SEQUENCES AMINO ACID AMIN ACID R-GROUP
ANALYSIS OF INDIVIDUAL SEQUENCES
SECONDARY STRUCTURE PREDICTION CHOU FASMAN Table 8 Chou & Fasman Secondary Structure Propensity of the Amino Acids Pα Pβ Pc Pα Pβ Pc A 1.42 0.83 0.75 M 1.45 1.05 0.5 C 0.7 1.19 1.11 N 0.67 0.89 1.44 D 1.01 0.54 1.45 P 0.57 0.55 1.88 E 1.51 0.37 1.12 Q 1.11 1.1 0.79 F 1.13 1.38 0.49 R 0.98 0.93 1.09 G 0.57 0.75 1.68 S 0.77 0.75 1.48 H 1 0.87 1.13 T 0.83 1.19 0.98 I 1.08 1.6 0.32 V 1.06 1.7 0.24 K 1.16 0.74 1.1 W 1.08 1.37 0.45 L 1.21 1.3 0.49 Y 0.69 1.47 0.84
SECONDARY STRUCTURE PREDICTION GOR METHOD
SECONDARY STRUCTURE PREDICTION PHD
SECONDARY STRUCTURE PREDICTION TABLE OF PERFORMANCES 75 70 65 60 55 50 45 CF GOR I LIM LEVIN PTIT JASEP7 GOR III ZHANG PHD Scores (%)
PAIRWISE SEQUENCE COMPARISON DOT PLOTS
PAIRWISE SEQUENCE COMPARISON DOT PLOTS UNFILTERED FILTERED
PAIRWISE SEQUENCE COMPARISON SEQUENCE ALIGNMENT GATTATACCA GATTACA GATTATACCA GATTA---CA GAP OF LENGTH 3 INSERTIONS AND DELETIONS ( INDEL ) ARE REPRESENTED BY GAPS IN ALIGNMENTS
PAIRWISE SEQUENCE COMPARISON WHAT ABOUT THIS: GCTACTAGTTCGCTTAGC GCTACTAGCTCTAGCGCGTATAGC WHICH ONE?? GCTACTAG-T-T--CGC-T-TAGC GCTACTAGCTCTAGCGCGTATAGC GCTACTAGTT------CGCTTAGC GCTACTAGCTCTAGCGCGTATAGC
PAIRWISE SEQUENCE COMPARISON NEEDLEMAN/WUNSCH ALGORITHM (EXAMPLE WITH NO PENALTY) SEQUENCE #1: GAATTCAGTTA; M = 11 (LETTERS) SEQUENCE #2: GGATCGA; N = 7 (LETTERS) SCORING SCHEME: Si,j = 1 IF POS I OF #1 IS THE SAME AS POS J OF #2 Si,j = 0 IF MISMATCH SCORE w = 0 (GAP PENALTY) STEPS: INITIALIZATION MATRIX FILL TRACEBACK
PAIRWISE SEQUENCE COMPARISON SEQUENCE #1: GAATTCAGTTA; M = 11 (LETTERS) SEQUENCE #2: GGATCGA; N = 7 (LETTERS) INITIALIZATION
PAIRWISE SEQUENCE COMPARISON SEQUENCE #1: GAATTCAGTTA; M = 11 (LETTERS) SEQUENCE #2: GGATCGA; N = 7 (LETTERS) MATRIX FILL FOR EACH POSITION IN THE MATRIX, Mi,j IS DEFINED TO BE THE MAXIMUM SCORE AT THE POSITION i,j: Mi,j = MAX[ Mi-1, j-1 +Si,j (match/mismatch in the diagonal), Mi,j-1 + w (gap in sequence #1), Mi-1,j + w (gap in sequence #2)]
PAIRWISE SEQUENCE COMPARISON SEQUENCE #1: GAATTCAGTTA; M = 11 (LETTERS) SEQUENCE #2: GGATCGA; N = 7 (LETTERS) TRACEBACK G_ A A T T C A G T T A G G _ A _ T C _ G A
PAIRWISE SEQUENCE COMPARISON GAP PENALTY GATCGCTACGCTCAGC A.C.C..C..T PERFECT SIMILARITY EVERYTIME!
PAIRWISE SEQUENCE COMPARISON ALIGNMENT GLOBAL ALIGNMENT G-ATES GRATED LOCAL ALIGNMENT DO NOT NEED TO ALIGN ALL THE BASES IN ALL SEQUENCES ALIGN BILLGATESLIKESCHEESE AND GRATEDCHEESE G-ATESLIKESCHEESE OR G-ATES & CHEESE GRATED-----CHEESE GRATED & CHEESE
SINGLE SEQUENCE HEURISTIC ALGORITHMS DATABASE SEARCHING > fasta myquery swissprot -ktup 2 search program query sequence sequence database optional parameters
SINGLE SEQUENCE HEURISTIC ALGORITHMS RESULTS THE BEST SCORES ARE The best scores are: initn init1 opt z-sc E(77110) gi 1706794 sp P49789 FHIT_HUMAN BIS(5'-ADENOSYL)- 996 996 996 1262.1 0 gi 1703339 sp P49776 APH1_SCHPO BIS(5'-NUCLEOSYL) 412 382 395 507.6 1.4e-21 gi 1723425 sp P49775 HNT2_YEAST HIT FAMILY PROTEI 238 133 316 407.4 5.4e-16 gi 3915958 sp Q58276 Y866_METJA HYPOTHETICAL HIT- 153 98 190 253.1 2.1e-07 gi 3916020 sp Q11066 YHIT_MYCTU HYPOTHETICAL 15.7 163 163 184 244.8 6.1e-07 gi 3023940 sp O07513 HIT_BACSU HIT PROTEIN 164 164 170 227.2 5.8e-06 gi 2506515 sp Q04344 HNT1_YEAST HIT FAMILY PROTEI 130 91 157 210.3 5.1e-05 gi 2495235 sp P75504 YHIT_MYCPN HYPOTHETICAL 16.1 125 125 148 199.7 0.0002 gi 418447 sp P32084 YHIT_SYNP7 HYPOTHETICAL 12.4 42 42 140 191.3 0.00058 gi 3025190 sp P94252 YHIT_BORBU HYPOTHETICAL 15.9 128 73 139 188.7 0.00082 gi 1351828 sp P47378 YHIT_MYCGE HYPOTHETICAL HIT- 76 76 133 181.0 0.0022 gi 418446 sp P32083 YHIT_MYCHR HYPOTHETICAL 13.1 27 27 119 165.2 0.017 gi 1708543 sp P49773 IPK1_HUMAN HINT PROTEIN (PRO 66 66 118 163.0 0.022 gi 2495231 sp P70349 IPK1_MOUSE HINT PROTEIN (PRO 65 65 116 160.5 0.03 gi 1724020 sp P49774 YHIT_MYCLE HYPOTHETICAL HIT- 52 52 117 160.3 0.031 gi 1170581 sp P16436 IPK1_BOVIN HINT PROTEIN (PRO 66 66 115 159.3 0.035 gi 2495232 sp P80912 IPK1_RABIT HINT PROTEIN (PRO 66 66 112 155.5 0.057 gi 1177047 sp P42856 ZB14_MAIZE 14 KD ZINC-BINDIN 73 73 112 155.4 0.058 gi 1177046 sp P42855 ZB14_BRAJU 14 KD ZINC-BINDIN 76 76 110 153.8 0.072 gi 1169825 sp P31764 GAL7_HAEIN GALACTOSE-1-PHOSP 58 58 104 138.5 0.51 gi 113999 sp P16550 APA1_YEAST 5',5'''-P-1,P-4-TE 47 47 103 137.8 0.56 gi 1351948 sp P49348 APA2_KLULA 5',5'''-P-1,P-4-T 63 63 98 131.3 1.3 gi 123331 sp P23228 HMCS_CHICK HYDROXYMETHYLGLUTA 58 58 99 129.4 1.6 gi 1170899 sp P06994 MDH_ECOLI MALATE DEHYDROGENA 70 48 91 122.9 3.7 gi 3915666 sp Q10798 DXR_MYCTU 1-DEOXY-D-XYLULOSE 75 50 92 121.9 4.3 gi 124341 sp P05113 IL5_HUMAN INTERLEUKIN-5 PRECU 36 36 85 121.3 4.7 gi 1170538 sp P46685 IL5_CERTO INTERLEUKIN-5 PREC 36 36 84 120.0 5.5 gi 121369 sp P15124 GLNA_METCA GLUTAMINE SYNTHETA 45 45 90 118.9 6.3 gi 2506868 sp P33937 NAPA_ECOLI PERIPLASMIC NITRA 48 48 92 117.4 7.6 gi 119377 sp P10403 ENV1_DROME RETROVIRUS-RELATED 59 59 89 117.0 8 gi 1351041 sp P48415 SC16_YEAST MULTIDOMAIN VESIC 48 48 97 117.0 8 gi 4033418 sp O67501 IPYR_AQUAE INORGANIC PYROPHO 38 38 83 116.8 8.3
SINGLE SEQUENCE HEURISTIC ALGORITHMS FASTA (FAST-ALL) 1) FIND K-TUPLES IN THE TWO SEQUENCES 2) IDENTIFY THE 10 HIGHEST SCORING REGIONS 3) INTRODUCE GAPS 4) DETERMINE BEST SEGMENT OF SIMILARITY
SINGLE SEQUENCE HEURISTIC ALGORITHMS 1) FIND K-TUPS IN THE TWO SEQUENCES position 1 2 3 4 5 6 7 8 9 10 11 protein 1 n c s p t a..... protein 2..... a c s p r k position in offset amino acid protein A protein B pos A - posb ----------------------------------------------------- a 6 6 0 c 2 7-5 k - 11 n 1 - p 4 9-5 r - 10 s 3 8-5 t 5 - ----------------------------------------------------- Note the common offset for the 3 amino acids c,s and p A possible alignment is thus quickly found - protein 1 n c s p t a protein 2 a c s p r k
SINGLE SEQUENCE HEURISTIC ALGORITHMS 2) IDENTIFY THE 10 HIGHEST SCORING REGIONS
SINGLE SEQUENCE HEURISTIC ALGORITHMS 3) INTRODUCE GAPS 4) DETERMINE BEST SEGMENT OF SIMILARITY
SINGLE SEQUENCE HEURISTIC ALGORITHMS BLAST (BASIC LOCAL ALIGNMENT SEARCH TOOL) 1) COMPILE THE LIST OF POSSIBLE WORDS 2) SCAN DATABASE FOR EXACT MATCHING 3) SCAN THROUGH THE LIST AND TRY TO EXTEND IT
SINGLE SEQUENCE HEURISTIC ALGORITHMS http://www.ebi.ac.uk/ http://www.ncbi.nlm.nih.gov/
ALIGNMENT AND SEARCH STATISTICS ALIGNMENT SCORE SUM OF THE WEIGHTS OF EVERY PAIR OF THE ALIGNMENT Z-SCORE (STANDARD DEVIATION FROM THE MEAN) Z-SCORE = (s-m)/e s = INITIAL SCORE m = MEAN OF THE RANDOM SCORES e = DEVIATION OF THE RANDOM SCORES
ALIGNMENT AND SEARCH STATISTICS EXTREME VALUE DISTRIBUTION (EVD) DENSITY FUNCTION:
ALIGNMENT AND SEARCH STATISTICS EXPECT VALUE E = NUMBER OF DATABASE HITS YOU EXPECT TO FIND BY CHANCE E = Kmne -λs Number K = scale for search space λ = scale for scoring system S = bitscore = (λs - lnk)/ln2 m = effective length of query n = effective length of database Score
MULTIPLE SEQUENCE ALIGNMENT EVOLUTION ALIGNMENT NYLS NKYLS NFS NFLS N-YLS NKYLS N-F-S N-FLS +K -L Y F NYLS
MULTIPLE SEQUENCE ALIGNMENT SUM-OF-PAIRS SCORE S(x, x) = 1 S(x, y) = -1 S(x, -) = -2 S(-, -) = 0 A A G 1 A - T -2 C C C 1 G G G 1 T T T 1 A A A 1 C - - -2 G A - -1 A A T 1 T T T 1 A G A -1 S(A,G) = -1 S(SEQ1,SEQ2) -1-1 -1-2 1 1 1 1 1 1 1 1-2 0-1 -2-1 -1 1 1 1-1 = = = = = = = = = = = S(SEQ1,SEQ3) -1-5 3 3 3 3-4 -5-1 3-1 = 2 S(SEQ2,SEQ3) SUM-OF-PAIRS SCORE
MULTIPLE SEQUENCE ALIGNMENT DYNAMIC PROGRAMMING C G T - G -G T A - - - - A G FOR n SEQUENCES OF SIZE l: SPACE: O(ln) TIME: O(2 n )
MULTIPLE SEQUENCE ALIGNMENT PROFILE EXAMPLE s(a,1)p(a,1) + s(b,1)p(b,1) + s(c,1)p(c,1) + S(d,1)p(D,1) = S(A,1)
MULTIPLE SEQUENCE ALIGNMENT CLUSTAL 1) COMPUTE A SIMILARITY OF PAIRS MATRIX
MULTIPLE SEQUENCE ALIGNMENT CLUSTAL 2) COMPUTE A TREE
MULTIPLE SEQUENCE ALIGNMENT CLUSTAL 3) BUILD THE FINAL MULTIPLE ALIGNMENT
MULTIPLE ALIGNMENT AND DB SEARCHING PROFILE SEARCHING MULTIPLE SEQUENCE ALIGNMENT CONSTRUCTION OF A PROFILE COMPARAISON TO A DATABASE BEST SIMILARITIES FOUND
MULTIPLE ALIGNMENT AND DB SEARCHING HIDDEN MARKOV MODEL Emission Probabilities Transition probabilities
MULTIPLE ALIGNMENT AND DB SEARCHING HIDDEN MARKOV MODEL SCORING #1 - T G C T - - A G G vrs: #2 - A C A C - - A T C Regular Expression ([AT] [CG] [AC] [ACTG]* A [TG] [GC]): #1 = Member #2: Member HMM: #1 = Score of 0.0023% #2 Score of 4.7%
PROTEIN FAMILIES AND PROTEIN DOMAINS PROSITE DATABASE http://www.expasy.org/prosite/
PROTEIN FAMILIES AND PROTEIN DOMAINS PFAM http://www.sanger.ac.uk/software/pfam/
CONCLUSION HIGHTHROUGHPUT COMPUTATIONAL ANALYSIS TOOL A BLESSING A PROBLEM
REFERENCES Bioinformatics Sequence and Genome Analysis David W. Mount Cold Spring Harbor Laboratory Press Notes and Powerpoint: http://www.hammoutene.com/epfl/isbio/
Thank you for your attention!