Today, to accelerate research in the bio-pharma space, from the creation of treatments for diseases to the production of new synthetic biomaterials, we are announcing a new Vertex AI solution that demonstrates how to use Vertex AI Pipelines to run DeepMind’s AlphaFold protein structure predictions at scale.

Once a protein’s structure is determined and its role within the cell is understood, scientists can develop drugs that can modulate the protein function based on its role in the cell. DeepMind, an AI research organization within Alphabet, created the AlphaFold system to advance this area of research by helping data scientists and other researchers to accurately predict protein geometries at scale.

In 2020, in the Critical Assessment of Techniques for Protein Structure Prediction (CASP14) experiment, DeepMind presented a version of AlphaFold that predicted protein structures so accurately, experts declared the “protein-folding problem” solved. The next year, DeepMind open sourced the AlphaFold 2.0 system. Soon after, Google Cloud released a solution that integrated AlphaFold with Vertex AI Workbench to facilitate interactive experimentation. This made it easier for many data scientists to efficiently work with AlphaFold, and today’s announcement builds on that foundation.

Last week, AlphaFold took another significant step forward when DeepMind, in partnership with the European Bioinformatics Institute (EMBL-EBI), released predicted structures for nearly all cataloged proteins known to science. This release expands the AlphaFold database from nearly 1 million structures to over 200 million structures—and potentially increases our understanding of biology to a profound degree. Between this continued growth in the AlphaFold database and the efficiency of Vertex AI, we look forward to the discoveries researchers around the world will make.

In this article, we’ll explain how you can start experimenting with this solution, and we’ll also survey its benefits, which include offering lower costs through optimized selection of hardware, reproducibility through experiment tracking, lineage and metadata management, and faster run time through parallelization.

Background for running AlphaFold on Vertex AI

Generating a protein structure prediction is a computationally intensive task. It requires significant CPU and ML accelerator resources and can take hours or even days to compute. Running inference workflows at scale can be challenging—these challenges include optimizing inference elapsed time, optimizing hardware resource utilization, and managing experiments.Our new Vertex AI solution is meant to address these challenges.

To better understand how the solution addresses these challenges, let’s review the AlphaFold inference workflow:

1. Feature preprocessing. You use the input protein sequence (in the FASTA format) to search through genetic sequences across organisms and protein template databases using common open source tools. These tools include JackHMMER with MGnify and UniRef90, HHBlits with Uniclust30 and BFD, and HHSearch with PDB70. The outputs of the search (which consist of multiple sequence alignments (MSAs) and structural templates) and the input sequences are processed as inputs to an inference model. You can run the feature preprocessing steps only on a CPU platform. If you’re using full-size databases, the process can take a few hours to complete.
2. Model inference. The AlphaFold structure prediction system includes a set of pretrained models, including models for predicting monomer structures, models for predicting multimer structures, and models that have been fine-tuned for CASP. At inference time, you independently run the five models of a given type (such as monomer models) on the same set of inputs. By default, one prediction is generated per model when folding monomer models, and five predictions are generated per model when folding multimers. This step of the inference workflow is computationally very intensive and requires GPU or TPU acceleration.
3. (Optional) Structure relaxation. In order to resolve any structural violations and clashes that are in the structure returned by the inference models, you can perform a structure relaxation step. In the AlphaFold system, you use the OpenMM molecular mechanics simulation package to perform a restrained energy minimization procedure. Relaxation is also very computationally intensive, and although you can run the step on a CPU-only platform, you can also accelerate the process by using GPUs.

The Vertex AI solution

The AlphaFold batch inference with the Vertex AI solution lets you efficiently run AlphaFold inference at scale by focusing on the following optimizations:

• Optimizing inference workflow by parallelizing independent steps.
• Optimizing hardware utilization (and as a result, costs) by running each step on the optimal hardware platform. As part of this optimization, the solution automatically provisions and deprovisions the compute resources required for a step.
• Describing a robust and flexible experiment tracking approach that simplifies the process of running and analyzing hundreds of concurrent inference workflows.

The following diagram shows the architecture of the solution.

The solution encompasses the following:

A strategy for managing genetic databases. The solution includes high-performance, fully managed file storage. In this solution, Cloud Filestore is used to manage multiple versions of the databases and to provide high throughput and low-latency access.

An orchestrator to parallelize, orchestrate, and efficiently run steps in the workflow. Predictions, relaxations, and some feature engineering can be parallelized. In this solution, Vertex AI Pipelines is used as the orchestrator and runtime execution engine for the workflow steps.

Optimized hardware platform selection for each step. The prediction and relaxation steps run on GPUs, and feature engineering runs on CPUs. The prediction and relaxation steps can use multi-GPU node configurations. This is especially important for the prediction step because the memory usage is approximately quadratic with the number of residues. Therefore, predicting a large protein structure can exceed the memory of a single GPU device.

Metadata and artifact management. The solution includes management for running and analyzing experiments at scale. In this solution, Vertex AI Metadata is used to manage metadata and artifacts.

The basis of the solution is a set of reusable Vertex AI Pipelines components that encapsulate core steps in the AlphaFold inference workflow: feature preprocessing, prediction, and relaxation. In addition to those components, there are auxiliary components that break down the feature engineering step into tools, and helper components that aid in the organization and orchestration of the workflow.

The solution includes two sample pipelines: the universal pipeline and a monomer pipeline. The universal pipeline mirrors the settings and functionality of the inference script in the AlphaFold Github repository. It tracks elapsed time and optimizes compute resources utilization. The monomer pipeline further optimizes the workflow by making feature engineering more efficient. You can customize the pipeline by plugging in your own databases.

Next steps

To learn more and to try out this solution, check our GitHub repository, which contains the components and universal and monomer pipelines. The artifacts in the repository are designed so that you can customize them. In addition, you can integrate this solution into your upstream and downstream workflows for further analysis. To learn more about Vertex AI, visit our product page.

Acknowledgements

We would like to thank the following people for their collaboration: Shweta Maniar, Sampath Koppole, Mikhail Chrestkha, Jasper Wang, Alex Burdenko, Meera Lakhavani, Joan Kallogjeri, Dong Meng (NVIDIA), Mike Thomas (NVIDIA), and Jill Milton (NVIDIA).

Finally and most importantly, we would like to thank our Solution Manager Donna Schut for managing this solution from start to finish. This would not have been possible without Donna.

Who would have known that today technology would enable us with the ability to use machine learning to track vessel activity, and make pattern inferences to help address IUU (illegal, unreported, and unregulated) fishing activities. What’s even more noteworthy is that we now have the computing power to share this information publicly in order to enable fair and sustainable use of our ocean. 

An amazing group of humans at the nonprofit Global Fishing Watch took on this massive big data challenge and succeeded. You can immediately access their dynamic map on their website globalfishingwatch.org/map that is bringing greater transparency to fishing activity and supporting the creation and management of marine protected areas throughout the world.

In our second episode of our People and Planet AI series we were inspired by their ML solution to this challenge, and we built a short video and sample with all the relevant code you need to get started with building a basic time-series classification model in Google Cloud, and visualize it in an interactive  map. 

Architecture

These are the components used to build a model for this sample:

• Global Fishing Watch GitHub: where we got the data
• Apache Beam: (open source library) runs on Dataflow. 
• Dataflow: (Google’s data processing service) creates 2 datasets; 1 for training a model and the other to evaluate its results.
• TensorflowKeras: (high level API library) used to define a machine learning model, which we then train in Vertex AI.
• Vertex AI: (a platform to build, deploy, and scale ML models) we train and output the model.

Pricing and steps

The total cost to run this solution was less than $5. 

There are seven steps we went through with their approximate time and cost:

Why do we use a time series classification model? 

Vessels in the ocean are constantly moving, which creates distinctive patterns from a satellite view.

 We can train a model to recognize the shapes of a vessel’s trajectory. Large vessels are required to use the automatic identification system, or AIS. The GPS-like transponders  regularly broadcast a vessel’s maritime mobile service identity, or MMSI, and other critical information to nearby ships, as well as to terrestrial and satellite receivers. While AIS is designed to prevent collisions and boost overall safety at sea, it has turned out to be an invaluable system for monitoring vessels and detecting suspicious fishing behavior globally.

One tricky part is that the MMSI data location signal (which includes a timestamp, latitude, longitude, distance from port, and more) is not emitted at regular intervals. AIS broadcast frequency changes with vessel speed (faster at higher speeds), and not all AIS messages that are broadcast are received – terrestrial receivers require line-of-sight, satellites must be overhead, and high vessel density can cause signal interference. For example, AIS messages might be received frequently as a vessel leaves the docks and operates near shore, then less frequently as they move further offshore until satellite reception improves.  This is challenging for a machine learning model to interpret. There are too many gaps in the data, which makes it hard to predict.

A way to solve this is to normalize the data and generate fixed-sized hourly windows. Then the model can predict if the vessel is fishing or not fishing for each hour.

It could be hard to know if a ship is fishing or not by just looking at its current position, speed, and direction. So we look at the data from the past as well, looking at the future could also be an option if we don’t need to do real time predictions. For this sample, it seemed reasonable to look 24 hours into the past to make a prediction. This means we need at least 25 hours of data to make a prediction for a single hour (24 hours in the past + 1 current hour). But we could predict longer time sequences as well. In general, to get hourly predictions, we need (n+24) hours of data.

Options to deploy and access the model

For this sample specifically we used Cloud Run to host the model as a web app so that other apps can call it to make predictions on an ongoing basis; this is our favorite in terms of pricing if you need to access your model from the internet over an extended period of time (charged per prediction request). You can also host it directly from Vertex AI where you trained and built the model, just note there is an hourly cost for using those VMs even if they are idle. If you do not need to access the model over the internet, you can make predictions locally or download the model onto a microcontroller if you have an IoT sensor strategy.

Want to go deeper?

If you found this project interesting and would like to dive deeper either into the specifics of the thought process behind each step of this solution or even run through the code in your own project (or test project); we invite you to check out our interactive sample hosted on Colab, which is a free Jupyter notebook.  It serves as a guide with all the steps to run the sample, including visualizing the predictions on a dynamically moving map using an open source Python library called Folium. 

There’s no prior experience required! Just click “open in Colab” which is linked at the bottom of GitHub.

You will need a Google Cloud Platform project. If you do not have a Google Cloud project you can create one with the free $300 Google Cloud credit, you just need to ensure you set up billing, and later delete the project after testing the desired sample.

🌏🌎🌍 We hope to inspire you to build other beautiful climate-related solutions.

Over 2.14 billion people worldwide are expected to buy online this year, according to Statista. Online retail sales will account for 22% of all purchases by 2023. But in a competitive retail landscape, positive interactions can mean the difference between a sale and an abandoned shopping cart.

One of the leading global marketplaces – Kinguin.net is a haven for gamers. Their bustling ecommerce business conducts over 500,000 new transactions monthly. Users will encounter over 50,000 unique digital products, from video games, gift cards, in-game items to computer software and services. With over 10 million registered users, Kinguin improved their experience by helping users find items quickly and deliver service at scale.

Helping customers find what they want, fast

Because of Kinguin’s high volume of users—both buyers and sellers—and breadth of digital products, browsing and shopping can be challenging. “Customers shop online for choice and convenience, but it can sometimes be overwhelming. We want anyone who shops at Kinguin to find what they are looking for quickly and easily,” says Viktor Romaniuk Wanli, Kinguin CEO and Founder.

Today’s retailers know that creating personalized shopping experiences is crucial for establishing and maintaining customer loyalty. Kinguin discovered their users were getting a rather standard retail experience. They wondered how they could offer them a more tailored, personalized experience.

They knew product recommendations were a great way to personalize experiences because they help customers discover products that match their tastes and preferences. But it’s not that easy to recommend products. Various shifting factors make recommendations much more complex:

• Customer behavior. Understanding customers is tough. How do you recommend something to a cold start user who’s never been to your site before? What happens when their behavior changes?
• Omnichannel context. According to Harvard Business Review, 73% of all customers use many channels when they buy. What happens when they go from desktop to mobile or from social media shopping to a proprietary app?
• Product data challenges. How do you recommend new products within a large catalog of items? What if your product data has sparse labeling or unstructured metadata?

Data wasn’t a problem for Kinguin. They had data orders, history, wishlists, and could collect events based on their platform interactions. It was the machine learning model expertise they lacked. So rather than building their own solution, they determined it was more cost effective for them to find a reliable partner. It was also essential that the solution integrated easily with Kubernetes, which enabled their global network.

With these considerations in mind, they applied for the Google Recommendations AI beta program. Kinguin became the first gaming e-commerce platform in Europe to use Recommendations AI when it launched in 2020.

Pro gamer move: using a fully managed AI service 

Google Recommendations AI uses algorithms to deliver highly personalized suggestions tailored to a customer’s preferences. Google Cloud based these algorithms on the same research that powers models by YouTube search and Google Shopping. Algorithms are always being tuned and adjusted to focus on individuals themselves—not just items.

Many shopping AIs rely on manually provisioning infrastructure and training machine learning models. Instead, Recommendations AI’s deep learning models use item and user metadata to gain insights. It processes Kinguin’s thousands of products at scale, iterating in real time. First, Kinguin pieces together a customer’s history and shopping journey. Then, using Recommendations AI, they can serve up personalized products—even for long-tail products and cold-start users. 

By leveraging internal tools, Kinguin didn’t need to start implementation from scratch. After a few trial sessions with Google Cloud engineers, they got started right away. Due to the fast-paced nature of a marketplace—i.e., price changes, out-of-stock items—Kinguin needed their recommendations to be as close to real time as possible. They used internal event buses to stream events and their product catalog directly to the recommendations API.

Kinguin rolled out in high-traffic areas, including their home page, product page, and category pages. They analyzed heat maps and scroll maps to figure out where to test placements. They also experimented with different recommendation models such as “recently bought together” and “you may like.” Engineers also factored in where they were implementing the models. For example, the “others you might like” model would fit best on the homepage, while “frequently bought together” made sense at checkout.

Understanding how product recommendations influence financials is critical for demonstrating the impact of personalization. Using BigQuery, Kinguin could analyze different cost projection models. BigQuery helped them dig into specific financial data to understand their margins and revenue gains.

Playing to win: enhanced customer experience

Since adopting Recommendations AI, Kinguin has improved both customer experience and satisfaction. Search times have shortened by 20 seconds. Additionally, their average cart value has increased by 5 EUR. Conversion rates have quadrupled since the outset. Click-thru rates have doubled, increasing by 2.16 on product pages and 2.8 times on recommendations pages.

“Google Recommendations AI has helped us evolve our service, increase customer loyalty and satisfaction. It has also contributed to a significant rise in sales,” says Wanli. Kinguin is already thinking about other ways of enhancing user experiences with recommendations. Ideas include their checkout process, other landing pages, and email marketing.

Kinguin’s journey with Google Cloud shows how companies can leverage AI to optimize sales and deliver high-performing, low-latency recommendations to any customer touchpoint. 

Learn more about Recommendations AI and Google Cloud AI and machine learning solutions.

With the novel coronavirus COVID-19 sweeping across continents and fatalities climbing every day, it was only a matter of time before countries closed their borders to contain its spread.

In the wake of this decision, travel and tourism, the linchpins of many economies, were among the worst affected.

According to the United Nations World Tourism Organization (UNWTO), the COVID-19 pandemic caused a 22 percent fall in international tourist arrivals during the first quarter of 2020, and could see an annual decline of between 60 percent and 80 percent when compared with 2019.

The impact on the economy and to livelihoods that are dependent on tourism and hospitality has been significant. Prior to the pandemic, the outlook was quite different. Research by UNWTO in 2018 estimated that India would have 50 million outbound tourists by 2020.

Technology was also set to play a huge part in that growth. A Google Travel study showed that 74 percent of travellers were planning their trips on the Internet, and technologies like AI, IoT and VR were all set to be key trends this year.

Now, as borders slowly reopen and travel restrictions are gradually lifted, technology could once again be the game-changer. Manoj Sharma CTO, Cleartrip.com spoke to YourStory about the industry’s road to recovery and how technology will aid that journey.

Read the Full Story on YourStory