Profiles and different Arrangement arrangements.


43 views
Uploaded on:
Description
Substance. Characterizing profilePSSM by PSI-BLASTProfile HMMAligning profilesPSSM
Transcripts
Slide 1

Profiles and numerous Sequence arrangements Understanding Bioinformatics 9 th KIAS winter school Lee, Juyong

Slide 2

Contents Defining profile PSSM by PSI-BLAST Profile HMM Aligning profiles PSSM & Profile HMM Generate different succession arrangement Progressive Other strategies

Slide 3

What is Profile? Speak to general properties of the arrangement of sequences An arrangement of successions contains more data than a solitary grouping Environment is being viewed as Two sorts P osition S pecific S coring M atrix Profile Hidden Markov Model

Slide 4

Example PSSM

Slide 5

Position particular scoring network $> blastpgp - b 0 - j 3 - h 0.001 - d myDB –I mySEQ.fasta –Q myPSSM.mtx –o myMSA.bla

Slide 6

An arrangement of successions has more data Are K, I and S are significant? Are A & T are useless? K, I and S are profoundly saved! T at the 6th segment is likewise rationed 2 nd and 4 th sections don\'t indicate inclination K-IAS- - KAI-ST-K-I-ST-KRISS- - K-I-STI K-IAS-KAI-ST

Slide 7

Generating PSSM Log-chances score of amino corrosive an at position u Multiple succession arrangement Lack of data ought to be dealt with! Not Good ! On the off chance that an is not watched, m  - ∞

Slide 8

Generating PSSM (2) Pseudo-checks : part of amino corrosive an at position u : amino corrosive a conveyance α & β are scaling parameters

Slide 9

Generating PSSM (3) More reasonable pseudocounts Use substitution network data as opposed to arbitrary arrangement! Pseudo tally of amino corrosive a F : recurrence of amino corrosive b at u Formula utilized as a part of PSI-BLAST

Slide 10

Example of Pseudocount

Slide 11

PSI-BLAST is grouping DB looking system Goal : Find arrangement homologs! In the first place, perform consistent BLAST nearby pursuit Build PSSM taking into account the first round result Align arrangements against PSSM Update grouping arrangement! Do these iteratively!

Slide 12

Sequence Logo

Slide 13

Profile HMM Represent general property of an arrangement of groupings in view of Hidden Markov Model 0.4 0.1 0.6 0.5 0.7 0.4 0.2 0.7 0.3 0.6 Emit Amino corrosive

Slide 14

Profile HMM (2) KIA-S-K-AIST KI- - ST KIA-S-K-AIST D1 D2 D3 D4 Start M1 M2 M3 M4 END I0 I1 I2 I3 I4 A S K T I

Slide 15

Profile HMM (3) KIA-S-K-AIST KI- - ST KIAS-KI-ST D1 D2 D3 D4 Start M1 M2 M3 M4 END I0 I1 I2 I3 I4 I S T K

Slide 16

Estimate probabilities Transition likelihood between states Amino corrosive emanation likelihood

Slide 17

Profile HMM requires a great deal of information Many parameters to be prepared Transition probabilities ~ N seq * 9 Amino corrosive outflow probabilities ~ N seq * 20 For 100 buildup seq, ~3000 parameters to be tuned Generally no less than 20~30 related successions are required to construct precise profile HMM

Slide 18

Many conceivable ways! We have to score them… … QUERY : KRISS D1 D2 D3 D4  Start M1 M2 M3 M4 END Start M1 M2 M3 M4 END I0 I1 I2 I3 I4 I0 I1 I2 I3 I4 S R I S K R 

Slide 19

How to score an arrangement to profile HMM Two methods for assessing wellness of a grouping to profile HMM Through the Most plausible way Viterbi calculation Faster, less exact Consider every single conceivable way ! Forward ( Backward ) calculation Slower, more exact

Slide 20

Viterbi calculation Equivalent to the dynamic programming of pairwise arrangement

Slide 21

Forward calculation Consider all conceivable way ! Likelihood of emanating x i at state S u

Slide 22

Summary Profile  General property of an inquiry succession got from an arrangement of related groupings Position particular Scoring Matrix Profile Hidden Markov Model Can discover remote succession homolog Those can not be recognized by pairwise arrangement of arrangements

Slide 23

Aligning Profiles Comparing PSSM LAMA : no holes permitted, use Pearson relationship of scores Prof_sim : holes permitted, use amino corrosive appropriation at every section COMPASS : crevices permitted, psuedocounts are utilized as like PSI-BLAST

Slide 24

Aligning profile HMMs COACH, HHsearch are accessible Can discover exceptionally remote homologs Position subordinate hole scoring is conceivable

Slide 25

Multiple Sequence Alignment - MSA

Slide 26

Why MSA is troublesome? DP of Pairwise is simple and appropriate Only three cases If three groupings… … Seven cases… … For six arrangements… … 60TB memory required DP is Impossible  An A - An A V - V - V L - L

Slide 27

Methods to adjust successions Progressive technique Add a succession at once ClustalW, T-COFFEE, and so forth. Iterative technique Deletion, realigning steps are presented Prrp, DIALIGN, MUSCLE and so on

Slide 28

Order is vital! Case 1 Let\'s adjust the followings - D-G D-G-D  - G-G- - G-G D-G-G D-G-G- - Case 2 D G-G D-G-D 

Slide 29

Determine request ! Fabricate phylogenic tree taking into account all pairwise separation network

Slide 30

Which MSA is better? - Scoring plan Usually Sum of Pairs are utilized

Slide 31

Scores ClustalW Similar to plans for pairwise arrangement Employ deposit particular crevice opening

Slide 32

Scores (2) T-COFFEE Score if adjusted section is available in the Library Diverse arrangement Local & Global

Slide 33

Library Extension of T-COFFEE Different Weights for individual segments

Slide 34

Other techniques - DIALIGN Construct entire arrangement from ungapped nearby arrangements Find all ungapped arrangements and weight them ! Key Idea : pairwise arrangement can miss organically essential area

Slide 35

Other strategies - SAGA Genetic Algorithm Alignment  era Evolve through transformation & Crossover

Slide 36

Other techniques - MSACSA

Slide 37

Thank you!

Recommended
View more...