Abstract

Abstract
Author summaryCancer therapies often fail because tumor cells become resistant to treatment. One way to overcome resistance is by treating patients with a combination of two or more drugs. Some combinations may be more effective than when considering individual drug effects, a phenomenon called drug synergy. Computational drug synergy prediction methods can help to identify new, clinically relevant drug combinations. In this study, we developed several deep learning models for drug synergy prediction. We examined the effect of using different types of deep learning architectures, and different ways of representing drugs and cancer cell lines. We explored the use of biological prior knowledge to select relevant cell line features, and also tested data-driven feature reduction methods. We tested both precomputed drug features and deep learning methods that can directly learn features from raw representations of molecules. We also evaluated whether including genomic features, in addition to gene expression data, improves the predictive performance of the models. Through these experiments, we were able to identify strategies that will help guide the development of new deep learning models for drug synergy prediction in the future. One of the main obstacles to the successful treatment of cancer is the phenomenon of drug resistance. A common strategy to overcome resistance is the use of combination therapies. However, the space of possibilities is huge and efficient search strategies are required. Machine Learning (ML) can be a useful tool for the discovery of novel, clinically relevant anti-cancer drug combinations. In particular, deep learning (DL) has become a popular choice for modeling drug combination effects. Here, we set out to examine the impact of different methodological choices on the performance of multimodal DL-based drug synergy prediction methods, including the use of different input data types, preprocessing steps and model architectures. Focusing on the NCI ALMANAC dataset, we found that feature selection based on prior biological knowledge has a positive impact-limiting gene expression data to cancer or drug response-specific genes improved performance. Drug features appeared to be more predictive of drug response, with a 41% increase in coefficient of determination (R-2) and 26% increase in Spearman correlation relative to a baseline model that used only cell line and drug identifiers. Molecular fingerprint-based drug representations performed slightly better than learned representations-ECFP4 fingerprints increased R-2 by 5.3% and Spearman correlation by 2.8% w.r.t the best learned representations. In general, fully connected feature-encoding subnetworks outperformed other architectures. DL outperformed other ML methods by more than 35% (R-2) and 14% (Spearman). Additionally, an ensemble combining the top DL and ML models improved performance by about 6.5% (R-2) and 4% (Spearman). Using a state-of-the-art interpretability method, we showed that DL models can learn to associate drug and cell line features with drug response in a biologically meaningful way. The strategies explored in this study will help to improve the development of computational methods for the rational design of effective drug combinations for cancer therapy.

Abstract
We propose Soteria, a system for distributed privacy-preserving Machine Learning (ML) that leverages Trusted Execution Environments (e.g. Intel SGX) to run code in isolated containers (enclaves). Unlike previous work, where all ML-related computation is performed at trusted enclaves, we introduce a hybrid scheme, combining computation done inside and outside these enclaves. The conducted experimental evaluation validates that our approach reduces the runtime of ML algorithms by up to 41%, when compared to previous related work. Our protocol is accompanied by a security proof, as well as a discussion regarding resilience against a wide spectrum of ML attacks.

Abstract
The adoption of third-party machine learning (ML) cloud services is highly dependent on the security guarantees and the performance penalty they incur on workloads for model training and inference. This paper explores security/performance trade-offs for the distributed Apache Spark framework and its ML library. Concretely, we build upon a key insight: in specific deployment settings, one can reveal carefully chosen non-sensitive operations (e.g. statistical calculations). This allows us to considerably improve the performance of privacy-preserving solutions without exposing the protocol to pervasive ML attacks. In more detail, we propose Soteria, a system for distributed privacy-preserving ML that leverages Trusted Execution Environments (e.g. Intel SGX) to run computations over sensitive information in isolated containers (enclaves). Unlike previous work, where all ML-related computation is performed at trusted enclaves, we introduce a hybrid scheme, combining computation done inside and outside these enclaves. The experimental evaluation validates that our approach reduces the runtime of ML algorithms by up to 41% when compared to previous related work. Our protocol is accompanied by a security proof and a discussion regarding resilience against a wide spectrum of ML attacks.

Abstract
The accurate prediction of biological age can bring important benefits in promoting therapeutic and behavioural strategies for healthy aging. We propose the development of age prediction models using multi-modal datasets, including transcriptomics, methylation and histological images from lung tissue samples of 793 human donors. From a technical point of view this is a challenging problem since not all donors are covered by the same data modalities and the datasets have a very high feature dimensionality with a relatively smaller number of samples. To fairly compare performance across different data types, we’ve created a test set including donors represented in each modality. Given the unique characteristics of the data distribution, we developed gradient boosting tree and convolutional neural network models for each dataset. The performance of the models can be affected by several covariates, including smoking history, and, most importantly, by a skewed distribution of age. Data-centric approaches, including feature engineering, feature selection, data stratification and resampling, proved fundamental in building models that were optimally adapted for each data modality, resulting in significant improvements in model performance for imbalanced regression. The models were then applied to the test set independently, and later combined into a multi-modal ensemble through a voting strategy, predicting age with a median absolute error of 4 years. Even if prediction accuracy remains a challenge, in this work we provide insights to address the difficulties of multi-modal data integration and imbalanced data prediction. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Abstract
Conhecer o perfil das mulheres brasileiras que atuam em Processamento de Linguagem Natural (PLN) é um importante passo para o desenvolvimento de políticas e programas que visem aumentar a inclusão e a diversidade nessa área. Este é o primeiro trabalho realizado no Brasil com este fim. A partir de dados coletados via consulta pública, Lattes e Linkedin, notou-se que o perfil é de uma formação em computação ou linguística, atuando em empresas ou universidades, mas com pouca diversidade étnica e aparente dificuldade em conciliar vida profissional e maternidade. Analisando mais especificamente o grupo “Brasileiras em PLN” constatou-se uma expressiva capacidade de publicação e orientação, mas ainda uma baixa colaboração entre nossas integrantes.