Generative Deep Learning 2nd Edition Pdf Free Download

Generative AI is the hottest topic in tech. This practical book teaches machine learning engineers and data scientists how to use TensorFlow and Keras to create impressive generative deep learning models from scratch, including variational autoencoders (VAEs), generative adversarial networks (GANs), Transformers, normalizing flows, energy-based models, and denoising diffusion models.

The book starts with the basics of deep learning and progresses to cutting-edge architectures. Through tips and tricks, you'll understand how to make your models learn more efficiently and become more creative.

Download Zip 🔥 https://tiurll.com/2y4ycr 🔥

The emergence of the 2019 novel coronavirus (COVID-19), for which there is no vaccine or any known effective treatment created a sense of urgency for novel drug discovery approaches. One of the most important COVID-19 protein targets is the 3C-like protease for which the crystal structure is known. Most of the immediate efforts are focused on drug repurposing of known clinically-approved drugs and virtual screening for the molecules available from chemical libraries that may not work well. For example, the IC50 of lopinavir, an HIV protease inhibitor, against the 3C-like protease is approximately 50 micromolar, which is far from ideal. In an attempt to address this challenge, on January 28th, 2020 Insilico Medicine decided to utilize a part of its generative chemistry pipeline to design novel drug-like inhibitors of COVID-19 and started generation on January 30th. It utilized three of its previously validated generative chemistry approaches: crystal-derived pocked-based generator, homology modelling-based generation, and ligand-based generation. Novel druglike compounds generated using these approaches were published at www.insilico.com/ncov-sprint/. Several molecules will be synthesized and tested using the internal resources; however, the team is seeking collaborations to synthesize, test, and, if needed, optimize the published molecules.

The application of multiple omics technologies in biomedical cohorts has the potential to reveal patient-level disease characteristics and individualized response to treatment. However, the scale and heterogeneous nature of multi-modal data makes integration and inference a non-trivial task. We developed a deep-learning-based framework, multi-omics variational autoencoders (MOVE), to integrate such data and applied it to a cohort of 789 people with newly diagnosed type 2 diabetes with deep multi-omics phenotyping from the DIRECT consortium. Using in silico perturbations, we identified drug-omics associations across the multi-modal datasets for the 20 most prevalent drugs given to people with type 2 diabetes with substantially higher sensitivity than univariate statistical tests. From these, we among others, identified novel associations between metformin and the gut microbiota as well as opposite molecular responses for the two statins, simvastatin and atorvastatin. We used the associations to quantify drug-drug similarities, assess the degree of polypharmacy and conclude that drug effects are distributed across the multi-omics modalities.

Drug-response patterns in individuals with complex disease, such as type 2 diabetes (T2D), are intricate. Multiple organs and confounders are typically involved including comorbidities and polypharmacy1,2. Conversely, treatment with one or more drugs and the associated polypharmacy effects can have considerable impact on the molecular profile of the individual; however, such changes are still largely unknown3. The increasing availability of deep phenotyping and multi-omics screening has proven to be beneficial in the characterization of T2D and other diseases4,5,6,7, and offer the opportunity to gain mechanistic insights on the action of drugs on disease processes.

Here we investigate vertical data integration, where multiple omics datasets have been generated for the same samples. Challenges that must be overcome include integration of data across multiple continuous and discrete data modalities, efficient handling of missing data or even large missing parts of specific data types, differences in dimensionality, modality-specific noise and how to extract associations across data modalities9,10,11. There are several strategies for vertical integration of multi-modal datasets, such as element-wise addition of one dataset at a time, learning individual representations for each dataset before fusion, or multi-dimensional fusion where representations are learned from the input data altogether9,12,13,14. Examples are multi-omics factor analysis (MOFA), iCluster, and data integration analysis for biomarker discovery using latent components (DIABLO) implemented in mixOmics, which can integrate multiple modalities11,14,15,16. However, these methods primarily focus on discovering factors or latent variables that can be used for visualization, clustering, or prediction of disease.

We have previously developed a deep-learning framework on the basis of variational autoencoders (VAE)17,18 for integration and binning of large amounts of unstructured metagenomics data19. Specifically, a VAE is based on deep neural networks and learns to transform high-dimensional data into a lower-dimensional space, termed a latent representation. During this process the two networks of the VAE learn the structure of input data and associations between the input variables. In our previous study, we found that the VAE could learn to integrate two datasets without any prior knowledge or statistical model19. Similarly, others have shown the capabilities of VAEs as integrative models for extracting the underlying signal in data for improving clustering and prediction12,20,21,22,23, as well as for handling large proportions of missing data24. We, therefore, speculated that such a model could be used to integrate even deeper cohort-level multi-omics datasets. While previous studies have primarily focused on stratifying patients using the underlying latent representation22,25,26 we were also interested in whether we could acquire insights into the complex relationships that the network learns through data integration.

For this purpose, we exploited that the decoder of the VAE is a generative model. Thus, the final trained decoder will be able to generate new examples of data from the learned latent distribution. On the basis of this principle, a variety of generative models have been used to generate new examples of data, such as single-cell RNA data and artificial human chromosomes27,28. Additionally, when combined with Bayesian decision theory they have been used for analysis of single-cell RNA data on the basis of variational inference29,30,31. Generative models also allow investigation of the effect that a virtual perturbation of the input data will have on the generated examples. For instance, Yeo et al. trained a generative model on single-cell RNA time-series data and then perturbed the input data to identify the effect of the perturbation on the output of the generative model32. Similarly, a recent study used the generative model of a VAE trained on protein evolutionary data to predict the effect that genetic variants have on the fitness of human proteins33. For our multi-modal data, we hypothesized that the generative ability of the VAE would allow us to identify associations between, for example, patient exposures and omics features.

We therefore developed a framework that is based on VAEs that we applied to a cohort of 789 people with newly diagnosed T2D with extensive multi-omics characterization. These modalities included genomics, transcriptomics, proteomics, metabolomics, and microbiomes as well as data on medication, diet questionnaires, and clinical measurements. Our method was able to integrate multi-omics data with clinical and categorical data and was resistant to systematic biases in the data as well as large amounts of missing data. Using an ensemble of generative VAE models, feature perturbation, univariate statistical methods, and Bayesian decision theory we identify cross omics associations. We compared the drug multi-omics profiles and showed that different drugs are associated with unique clinical and molecular profiles. Our method, multi-omics variational autoencoders (MOVE) is freely available, easily scalable, can integrate any number of categorical and continuous datasets, and able to identify features to multi-omics associations.

Previous work to stratify the newly diagnosed T2D individuals from this cohort used 32 clinical features to identify four archetypes representing different T2D subtypes7. In addition, they used metformin status of the individuals to investigate if the subgroups were confounded by metformin treatment and found no significant impact on the clusters and their multi-omics correlations. In contrast to their work, we added medication data on 19 additional drugs and used all data as input to our unsupervised deep-learning model allowing the model to learn from all inputs simultaneously. Thus, we were able to identify associations between the drugs and multi-omics data, including for metformin indicating the importance of vertical integration.

Finally, we emphasize that our approach is, of course, not limited to drug associations; in principle, all the omics data could be assessed for associations across the datasets. We therefore believe that our generative method opens new possibilities in big multi-omics data analysis for discoveries of potential new biomarkers, carrying out gedankenexperiments, and investigating potential direct effects of drugs in high dimensionality molecular data that leads to testable hypotheses.

Much scientific enquiry across disciplines is founded upon a mechanistic treatment of dynamic systems that ties form to function. A highly visible instance of this is in molecular biology, where characterizing macromolecular structure and dynamics is central to a detailed, molecular-level understanding of biological processes in the living cell. The current computational paradigm utilizes optimization as the generative process for modeling both structure and structural dynamics. Computational biology researchers are now attempting to wield generative models employing deep neural networks as an alternative computational paradigm. In this review, we summarize such efforts. We highlight progress and shortcomings. More importantly, we expose challenges that macromolecular structure poses to deep generative models and take this opportunity to introduce the structural biology community to several recent advances in the deep learning community that promise a way forward. e24fc04721