A Talk on Protein Homology Detection by HMM-HMM comparisons[1] - - PowerPoint PPT Presentation

▶

Sep 21, 2022 460 likes •634 views

A Talk on Protein Homology Detection by HMM-HMM comparisons[1] Sding, J Qing Ye Department of Computer Science University of Illinois Urbana-Champaign March 15, 2017 1 / 16 Qing Ye qingye3@illinois.edu Protein homology detection by

SLIDE 1

A Talk on Protein Homology Detection by HMM-HMM comparisons[1] Söding, J

Qing Ye Department of Computer Science University of Illinois Urbana-Champaign March 15, 2017

Qing Ye qingye3@illinois.edu Protein homology detection by HMM-HMM comparison 1/16

1/16

SLIDE 2

Introduction

Homology detection and sequence alignment are essential for

protein structure prediction, function prediction and evolution.

HHSearch
Generalized sequence alignment with a profile HMM to pairwise

HMM alignment

Included predicted secondary structure to further improve the

sensitivity

The higher sensitivity also leads to increased alignment quality

Qing Ye qingye3@illinois.edu Protein homology detection by HMM-HMM comparison 2/16

2/16

SLIDE 3

Log-sum-of-odds score

Generalize to pairwise HMM alignment

Note that the equation is very similar to that of

sequence-HMM alignment

Here, p(x1, ...xL|Null) is simply the fixed amino acid

background frequency SLSO = P(x1, ..., xL|co-emission on path) P(x1, ..., xL|Null)

Qing Ye qingye3@illinois.edu Protein homology detection by HMM-HMM comparison 3/16

3/16

SLIDE 4

Notations

qi(k) and pj(k) are the emission probability at i and jth

position

f(a) is the background frequency
Ptr is the product of all transition probability through p and q.

SLSO =

∑

k:XkYk=MM

Saa(qi(k), pk(k)) + log Ptr Saa(qi, pj) = log

∑

a=1

qi(a)pj(a) f(a)

Qing Ye qingye3@illinois.edu Protein homology detection by HMM-HMM comparison 4/16

4/16

SLIDE 5

Pairwise alignment of HMM

Qing Ye qingye3@illinois.edu Protein homology detection by HMM-HMM comparison 5/16

5/16

SLIDE 6

Pairwise alignment of HMM

State Transfer

Match and Insert states emit amino acid while Delete state

does not

Match and Insert cannot be aligned with Delete
Delete can only aligned to Delete state or Gap state
Gap denotes that the column aligned with the Gap does not have

homologous partner

Qing Ye qingye3@illinois.edu Protein homology detection by HMM-HMM comparison 6/16

6/16

SLIDE 7

Pairwise alignment of HMM

SMM(i, j) = Saa(qi, pj) + max

                  

SMM(i − 1, j − 1) + log [qi−1(M, M)pj−1(M, M)] SMI(i − 1, j − 1) + log [qi−1(M, M)pj−1(I, M)] SIM(i − 1, j − 1) + log [qi−1(I, M)pj−1(M, M)] SDG(i − 1, j − 1) + log [qi−1(D, M)pj−1(M, M)] SGD(i − 1, j − 1) + log [qi−1(M, M)pj−1(D, M)]

Qing Ye qingye3@illinois.edu Protein homology detection by HMM-HMM comparison 7/16

7/16

SLIDE 8

Pairwise alignment of HMM

SMI(i, j) = max

  

SMM(i − 1, j) + log [qi−1(M, M)pj(M, I)] SMI(i − 1, j) + log [qi−1(M, M)pj(I, I)] SDG(i, j) = max

  

SMM(i − 1, j) + log [qi−1(M, D)] SDG(i − 1, j) + log [qi−1(D, D)]

Qing Ye qingye3@illinois.edu Protein homology detection by HMM-HMM comparison 8/16

8/16

SLIDE 9

A few tricks are added to improve the performance

Saa offset =

⇒ produce shorter alignment

Sequence weighting and pseudo-count =

⇒ alignment of sequences of several subdomains transform into profile as if the alignment was cut into subdomains first and then calcualte the profile

Scoring correlation =

⇒ to expect high column scores to appear in clusters

Qing Ye qingye3@illinois.edu Protein homology detection by HMM-HMM comparison 9/16

9/16

SLIDE 10

Scoring secondary structure

Compare against actual secondary structure

σ is a DSSP state
ρ is a PSIPred state
c is the confidence value

MSS(σ; ρ, c) = log P(σ; ρ, c) P(σ)P(ρ, c)

Qing Ye qingye3@illinois.edu Protein homology detection by HMM-HMM comparison 10/16

10/16

SLIDE 11

Compare against predicted secondary structure

Runs over all DSSP state

MSS(ρq

i, cq i; ρq j, cq j) = log

∑

P(ρq

i, cq i|σ)P(ρq j, cq j|σ)

P(ρq

i, cq i)P(ρq j, cq j)

P(σ)

Qing Ye qingye3@illinois.edu Protein homology detection by HMM-HMM comparison 11/16

11/16

SLIDE 12

Results

Qing Ye qingye3@illinois.edu Protein homology detection by HMM-HMM comparison 12/16

12/16

SLIDE 13

Results

Qing Ye qingye3@illinois.edu Protein homology detection by HMM-HMM comparison 13/16

13/16

SLIDE 14

Results

Qing Ye qingye3@illinois.edu Protein homology detection by HMM-HMM comparison 14/16

14/16

SLIDE 15

Results

Qing Ye qingye3@illinois.edu Protein homology detection by HMM-HMM comparison 15/16

15/16

SLIDE 16

Reference I

[1] J. Söding. Protein homology detection by hmm–hmm

comparison. Bioinformatics, 21(7):951, 2005.

Qing Ye qingye3@illinois.edu Protein homology detection by HMM-HMM comparison 16/16

16/16