Predicting and Understanding HIV-1 Resistance to Broadly - - PowerPoint PPT Presentation

▶
predicting and understanding hiv 1 resistance to broadly
SMART_READER_LITE
LIVE PREVIEW

Predicting and Understanding HIV-1 Resistance to Broadly - - PowerPoint PPT Presentation

Jul 01, 2023 384 likes •720 views

Predicting and Understanding HIV-1 Resistance to Broadly Neutralizing Antibodies Anna Feldmann Max Planck Institute for Informatics Motivation HIV-1 drug target space is limited Drug resistance emergence under HAART Consistent

slide-1
SLIDE 1

Predicting and Understanding HIV-1 Resistance to Broadly Neutralizing Antibodies

Anna Feldmann

Max Planck Institute for Informatics

slide-2
SLIDE 2

1 Anna Feldmann

Motivation

De Clercq et al., Nature Reviews Drug Discovery 2007

  • HIV-1 drug target space is limited
  • Drug resistance emergence under HAART
  • Consistent change of treatment for

chronic patients Need for new drug targets

slide-3
SLIDE 3

2 Anna Feldmann

Motivation

De Clercq et al., Nature Reviews Drug Discovery 2007

slide-4
SLIDE 4

2 Anna Feldmann

Motivation

Burton et al., Nature Reviews Immunology 2002 De Clercq et al., Nature Reviews Drug Discovery 2007

(new) target: envelope spike treatment option: broadly neutralizing antibodies (bNAbs)

slide-5
SLIDE 5

3 Anna Feldmann

Broadly Neutralizing Antibodies

Burton et al., Science 2012

45-46G54W (CD4bs)

(Diskin et al., 2011)

PGT128 (V3-loop)

(Walker et al., 2011; Mouquet et al., 2012)

PG9 (V1/V2-loop)

(Walker et al., 2009)

Glycans 4E10 (MPER)

(Cardoso et al., 2005)

slide-6
SLIDE 6

3 Anna Feldmann

Broadly Neutralizing Antibodies

Burton et al., Science 2012

45-46G54W (CD4bs)

(Diskin et al., 2011)

PGT128 (V3-loop)

(Walker et al., 2011; Mouquet et al., 2012)

PG9 (V1/V2-loop)

(Walker et al., 2009)

Glycans 4E10 (MPER)

(Cardoso et al., 2005)

Developed in 10-30% of infected patients HIV-1 vaccine research: HIV-1 treatment: BUT: too little too late Induce bNAb development bNAb immunotherapy

slide-7
SLIDE 7

4 Anna Feldmann

bNAbs for HIV-1 Treatment

slide-8
SLIDE 8

4 Anna Feldmann

bNAbs for HIV-1 Treatment

slide-9
SLIDE 9

4 Anna Feldmann

bNAbs for HIV-1 Treatment

slide-10
SLIDE 10

5 Anna Feldmann

Challenges Goal

  • 1. Resistance to bNAbs

(Will it work?)

  • 2. Optimal Personalized Treatment

(Which will work best?) Given HIV-1 variants of the patient and a bNAb Predict susceptibility or resistance

bNAbs for HIV-1 Treatment

slide-11
SLIDE 11

6 Anna Feldmann

Data

Neutralization assay data covering 4 major epitopes: CD4bs:

VRC01, VRC-PG04, NIH45-46, 3BNC117

gp41-gp120:

35O22

V3-loop:

PGT128, PGT121, 10-1074, 10-996

V1/V2-loop:

PG9, PG16

Doria-Rose et al., J.Virol. 2009; Mouquet et al., PNAS 2012; Huang et al., Nature 2014

slide-12
SLIDE 12

6 Anna Feldmann

Data

Neutralization assay data covering 4 major epitopes: CD4bs:

VRC01, VRC-PG04, NIH45-46, 3BNC117

gp41-gp120:

35O22

V3-loop:

PGT128, PGT121, 10-1074, 10-996

V1/V2-loop:

PG9, PG16

115-230 envelope (Env) sequences with corresponding IC50 values per bNAb

Doria-Rose et al., J.Virol. 2009; Mouquet et al., PNAS 2012; Huang et al., Nature 2014

slide-13
SLIDE 13

6 Anna Feldmann

Building Prediction Model

  • 1. Learning
slide-14
SLIDE 14

6 Anna Feldmann

Building Prediction Model

  • 1. Learning
slide-15
SLIDE 15

6 Anna Feldmann

Building Prediction Model

  • 1. Learning
  • resistant (–),

if IC50 ≥ 50μg/mL

  • susceptible (+),

if IC50 < 50μg/mL Binarize label into

slide-16
SLIDE 16

6 Anna Feldmann

Building Prediction Model

  • 1. Learning
  • resistant (–),

if IC50 ≥ 50μg/mL

  • susceptible (+),

if IC50 < 50μg/mL Binarize label into

slide-17
SLIDE 17

6 Anna Feldmann

Building Prediction Model

  • 1. Learning
  • 2. Predicting
slide-18
SLIDE 18

7 Anna Feldmann

Prediction Performance

  • Model performance

was tested in a 10 - times nested cross- validation

  • Overall high

prediction performance (up to 0.84 AUC)

  • Classifiers for the

same epitope achieve similar performances

slide-19
SLIDE 19

7 Anna Feldmann

Prediction Performance

Questions: Can we interpret the models? Can we interpret the classification result?

slide-20
SLIDE 20

8 Anna Feldmann

Understanding the Classifier

Learnt discriminant positions of the classifiers

susceptible resistant

slide-21
SLIDE 21

Anna Feldmann

Understanding the Result

Residues of the test sequence that contributed the most (strongest 5%) to the classification result of the PG9 classifier.

9

AA susceptible aa resistant

slide-22
SLIDE 22

10 Anna Feldmann

bNAbs for HIV-1 Treatment

Challenges Goal

  • 1. Resistance to bNAbs

(Will it work?)

  • 2. Optimal Personalized Treatment

(Which will work best?) Predict susceptibility or resistance Given HIV-1 variants of the patient and a bNAb

slide-23
SLIDE 23

10 Anna Feldmann

bNAbs for HIV-1 Treatment

Challenges Goal

  • 1. Resistance to bNAbs

(Will it work?)

  • 2. Optimal Personalized Treatment

(Which will work best?) Predict the corresponding IC50 value Given HIV-1 variants of the patient and a bNAb

slide-24
SLIDE 24

11 Anna Feldmann

Building Regression Models

Setup:

  • Same input data
  • Instead of binarization,

log transformation used

  • Instead of classification, the

corresponding IC50 value is predicted using support vector regression

slide-25
SLIDE 25

11 Anna Feldmann

Building Regression Models

Setup:

  • Same input data
  • Instead of binarization,

log transformation used

  • Instead of classification, the

corresponding IC50 value is predicted using support vector regression Positive correlations of 0.3 – 0.5 for all bNAbs apart from 35O22 Result:

slide-26
SLIDE 26

12 Anna Feldmann

Continuous Drift Towards Resistance

Studied population:

  • 40 Caucasian men

having sex with men, subtype B

  • similar distribution of

viral loads and CD4-T cell counts

  • b12, VRC01, VRC03,

NIH45-46G54W, PG9, PG16, PGT121, PGT128, PGT145

  • Over 20 years

(1987–1991/ 1996–2000/ 2006–2010)

  • French ANRS PRIMO

and SEROCO cohorts

slide-27
SLIDE 27

Anna Feldmann

Questions: Only for subtype B? Does it hold for global viral population? What about other time periods?

Studied population:

  • 40 Caucasian men

having sex with men, subtype B

  • similar distribution of

viral loads and CD4-T cell counts

  • b12, VRC01, VRC03,

NIH45-46G54W, PG9, PG16, PGT121, PGT128, PGT145

  • Over 20 years

(1987–1991/ 1996–2000/ 2006–2010)

  • French ANRS PRIMO

and SEROCO cohorts

12

Continuous Drift Towards Resistance

slide-28
SLIDE 28

Anna Feldmann

Time Analysis over LANL Env Seqs

Setup:

  • ~36.000 Env Seqs from

LANL, different subtypes

  • Time covered:

1981-2013 Paper vs our time partitioning

  • Predicted IC50 value

using support vector regression models

1981 1987 1992 1996 2000 2006 2013 2010 before ART ART cocktail (NRTIs) HAART cocktail with PIs LPV/r Maraviroc/ Raltegravir ART monotherapy

13

Doria-Rose et al., J.Virol. 2009; Mouquet et al., PNAS 2012; Huang et al., Nature 2014

slide-29
SLIDE 29

Anna Feldmann

Time Analysis over LANL Env Seqs

14

  • Continuous trend towards resistance

for all antibodies but PG9 and PG16 (Bonferroni correction threshold 0.05/22=~0.002, umbrella test)

  • Considering non-B subtype (vs B):

similar trend, but PGT121, PGT128 not significant anymore (Bonferroni correction threshold 0.05/22=~0.002, umbrella test)

  • Over the whole available time period

l

  • g

( I C 5 ) Time periods NIH45-46 PG9

slide-30
SLIDE 30

Anna Feldmann

Conclusion

  • Well performing classification models for HIV-1 resistance

to bNAbs

  • Reliable classifiers identifying potential binding site

residues

  • Visualization of data relationships and motif logos improve

biological understanding of the classification result

  • Regression models provide more fine-grained predictions
  • Useful as recommendation device for bNAb combination

therapy

  • Extendable to new HIV-1 bNAbs or HCV bNAbs

15

slide-31
SLIDE 31

Anna Feldmann

Thanks to ...

Max Planck Institute for Informatics, Saarbrücken

  • Thomas Lengauer
  • Nico Pfeifer
  • Alejandro Pironti
  • Nora Speicher

and Rolf Kaiser

16

slide-32
SLIDE 32

Anna Feldmann

Thanks to ...

Max Planck Institute for Informatics, Saarbrücken

  • Thomas Lengauer
  • Nico Pfeifer
  • Alejandro Pironti
  • Nora Speicher

and Rolf Kaiser

you for listening! Questions?

… you for listening.

16