History-alignment Models for Bias-aware Prediction of Virological - - PowerPoint PPT Presentation

History-alignment Models for Bias-aware Prediction of Virological Response to HIV Combination Therapy

Jasm ina Bogojeska

Department of C

• mputational Biology and Applied

Algorithmics, Max-Planck Institute for Informatics AR E VIR – G enaF

• r – Meeting, 2012

Problem Setting

Bias-aware prediction of the

• utcome of combination therapies

given to HIV patients Develop methods that can deal with:

Evolving trends in treating patients over time Sparse, uneven therapy representation Different treatment backgrounds

• f the samples

History-Aware Models

Problem with clinical data:

Samples with different treatment backgrounds – uneven sample representation regarding level of therapy experience Only dominant viral strain sequenced – no information on latent virus population

Idea: Use treatment history information

Therapy sequence Pairwise similarity of therapy sequences

Adapt sequence alignment methods!

)} ( ) ( and ) ( ) ( | { ) ( t z t z z t patient patient start start r = ≤ =

Similarity of Therapy Sequences

Quantify pairwise therapy similarity

Use drug resistance mutations

Therapy sequence alignment

∑

∈

+

) , , , (

) ), , , , ( ( )) ( ), ( ( | | 1 max arg

D y T i i i i i

i i i i

y f l r r S D

h z x t t t w

w w w h z x t z

t

σ

γ

History Alignment Model

Train a separate model for each therapy sequence by using knowledge from similar therapy sequences

Sample weighted regularized logistic regression

• Consider the treatment backgrounds of the samples, the latent virus

population and the current therapy

• Account for the sparse representation of the different therapy-histories
• Account for the missing information on the latent virus population
• Account for the sparse, uneven therapy representation

loss function seq similarity

• Cluster the training data using the pairwise similarities of their corresponding therapy

sequences

• Apply the multi-task distribution matching method with clusters as tasks
• For each (target) cluster t:

Multi-class logistic regression Sample-weighted logistic regression

History Distribution Matching Model

History Distribution Matching Model

Added advantages

Address increasing imbalance in the representation of the effective and ineffective therapies over time

Independent estimation of the models for the effective and the ineffective therapies in the distribution matching step Time-oriented model selection

Address missing treatment history information

The model for each cluster also uses the data from other clusters with apropriately derived relevance weights

Data set training tuning test Sample count

3596 1634 1307

Success rate

69 % 79 % 83 %

Results (Treatment History)

• History-aware models achieve better predictions for treatment-experienced

patients

Results (Therapy Abundances)

• History-aware models achieve better predictions for rare therapies

Conclusions

Treatment history-aware methods

Information extracted from treatment history enhances the performance for therapy-experienced patients and for rare therapies History-alignment method

Patient-specific models that utilize detailed treatment history information

History distribution matching method

Address the increasing gap between the representation of successful and failing therapies over time Tackle the problem of missing treatment history information All this enhances the accuracy performance

Acknowledgements

Thomas Lengauer and the HIV group at MPI Rolf Kaiser and his group

Thank You!

Data

Viral genotype Current treatment Treatment history Label (success or failure)

0 1 0 0 1 …

• ccurrence of

resistance mutations 1 1 0 0 1 … 1 0 0 1 0 … drugs used in current treatment drugs used in all previous treatments 1 or -1

6336 labeled samples with different 638 combination therapies