Deep Learning für die Biowissenschaften: Einsatz von Deep Learning in Genomik, Biophysik, Mikroskopie und medizinischer Analyse

by Bharath Ramsundar

With much success already attributed to deep learning, this discipline has started making waves throughout science broadly and the life sciences in particular. With this practical book, developers and scientists will learn how deep learning is used for genomics, chemistry, biophysics, microscopy, medical analysis, drug discovery, and other fields.As a running case study, the authors focus on the problem of designing new therapeutics, one of science's greatest challenges because this practice ties together physics, chemistry, biology and medicine. Using TensorFlow and the DeepChem library, this book introduces deep network primitives including image convolutional networks, 1D convolutions for genomics, graph convolutions for molecular graphs, atomic convolutions for molecular structures, and molecular autoencoders.Deep Learning for the Life Sciences is ideal for practicing developers interested in applying their skills to scientific applications such as biology, genetics, and drug discovery, as well as scientists interested in adding deep learning to their core skills.

Genres: Science, Technical, Textbooks, Artificial Intelligence

Paperback

Published February 29, 2020

Reviews

[James Warren · Rating 3 out of 5]

This book has a distinctive "color by numbers" feel. It provides highly useful background information, but the provided examples have limited illustrative value. Reading this book is a nice overview, but it certainly will not prepare you to conduct similar analyses independently.

[Will Morgan · Rating 3 out of 5]

While this is an interesting intro to multiple areas of deep learning across bioinformatics, cheminformatics, and genomics, it has some flaws in that deepchem, the python package this book revolves around, needs some very specific versions of python, tensorflow, scikit learn, and more. It's common to be overwhelmed with warning labels about deprecations, or simply things not working.

While yes, it is possible to make a separate environment and run the book exercises there, it does raise complications for using deepchem with other packages and as part of a wider project.

I wonder if this package is still fully maintained, or if any parts have been abandoned, as some functionalities gone through in the book simply do not exist, even when the docs site is searched. This book makes deepchem seem exciting and full of potential until the point where you try to use it.

[Miguel Madero · Rating 4 out of 5]

Useful intro, it covers a very wide range of applications (most of life-sciences) and techniques (a good subset of Deep Learning). It has a high-level explanation of both the field and the potential DL techniques that could be used, the current state (benchmarks and real-world uses), limitations, and potential.

Chapter 1, 2, and 3 are good intros to both LifeSciences, DeepLearning, and Deechem (a library used throughout the book). Then each chapter focuses on a different area of LifeSciences. ML for Molecules is a great example of good high-level coverage.

Minor critiques: The last few chapters felt a bit rushed. The code samples could use some work and maybe a better setup with Jupiter notebooks or something similar so it's easier to follow along.

[Elmer Peramo · Rating 2 out of 5]

This is more like a survey paper than a cohesive textbook. There are some interesting chapters, but others are lacking in depth and detail. It's better for experienced deep learning practitioners who want to learn domains in chemistry, biophysics, and genomics than for biologists who need to study deep learning.

[Richard · Rating 5 out of 5]

I was looking for a summary of how life science data is represented in machine learning experiments. This is perfect for what I needed.

[Radu Angelescu · Rating 3 out of 5]

Nice

It's a nice book. I would have liked it to be more in depth but I guess it's meant to be for casual readers.

[Josua Naiborhu · Rating 4 out of 5]

interesting book to read for anyone who is interested in deep learning for the life sciences in the introductory level.