


Automatically identify the best molecules and reduce synthesis costs. MIT develops a molecular design decision-making algorithm framework
Editor | Ziluo
The use of AI to streamline drug discovery is exploding. Screen billions of candidate molecules for those that may have properties needed to develop new drugs. There are so many variables to consider, from material prices to the risk of error, that weighing the costs of synthesizing the best candidate molecules is no easy task, even when scientists use AI.
Here, MIT researchers developed SPARROW, a quantitative decision algorithm framework, to automatically identify the best molecular candidates, thereby minimizing synthesis costs while maximizing the likelihood that the candidate has the desired properties. The algorithm also identified the materials and experimental steps required to synthesize these molecules.
SPARROW takes into account the cost of synthesizing a batch of molecules at once, since multiple candidate molecules can often be derived from some of the same compounds. Furthermore, this unified approach enables access to critical information for molecular design, property prediction, and synthesis planning from online repositories and widely used AI tools.
In addition to helping pharmaceutical companies discover new drugs more efficiently, SPARROW can also be used to invent new agricultural chemicals or discover specialized materials for organic electronics.
Relevant research titled "An algorithmic framework for synthetic cost-aware decision making in molecular design" was published on "Nature Computational Science" on June 19.
Paper link: https://www.nature.com/articles/s43588-024-00639-y
"Selection of compounds is an art, and sometimes it is a very successful art. But given that we have all these models and prediction tools that provide information about how molecules might behave and be synthesized, we should use that information to guide the decisions we make," said Connor, corresponding author of the paper and an assistant professor in the Department of Chemical Engineering at MIT. Coley said.
Quantitative decision-making algorithm framework SPARROW
"Synthesis Planning And Rewards-based Route Optimization Workflow, SPARROW" is an algorithmic decision-making framework used to drive the design cycle.
Illustration: Overview of SPARROW and its role in the molecular design cycle. (Source: Paper)
This research builds on earlier problem formulations for the simultaneous selection of synthetic routes for multiple molecules, and the integration of product and process system design. Unlike traditional screening methods, SPARROW uses a multi-objective optimization criterion that balances cost and utility to prioritize molecules and their putative synthetic routes from a library of candidate molecules.
SPARROW generates reaction networks consisting of candidate target molecules and synthetic routes. By solving graph-based optimization problems, a set of molecules and synthetic routes can be screened to optimally balance cumulative synthetic cost and utility. In this context, utility measures the value of assessing a molecular property.
Appropriate measures of utility will vary at different stages of application and design. It may include predictions of molecular properties, uncertainties in these predictions, or the potential of new data points to improve structure-property relationships. A library of candidates must be provided to SPARROW with a corresponding reward indicating the utility associated with each candidate molecule.
Illustration: SPARROW’s problem statement. (Source: Paper)
The reward for choosing a molecule also depends on the success of the chosen reaction steps to synthesize that molecule. If a reaction step in the synthetic route of a candidate molecule fails, no information is gained. The researchers formalized this by maximizing the expected reward of selecting a candidate molecule, which can be expressed as its reward multiplied by the probability of successfully synthesizing the molecule.
Balancing cost and utility, the goal of SPARROW can be formalized as the expected reward of all selected goals divided by the cost of synthesizing all selected goals using the selected route.
Complex cost considerations
In a sense, whether scientists should synthesize and test a certain molecule comes down to a question of the cost of synthesis versus the value of the experiment. However, determining cost or value is a difficult problem in itself.
SPARROW addresses this challenge by taking into account the shared intermediate compounds involved in synthesizing a molecule and incorporating this information into its cost vs. value function.
“When you think about the optimization problem of designing a batch of molecules, the cost of adding new structures depends on the molecules you’ve already chosen,” Coley said.
The framework also takes into account factors such as the cost of the starting materials, the number of reactions involved in each synthesis route, and the likelihood of those reactions being successful on the first try.
To use SPARROW, scientists provide a set of molecular compounds they are considering testing, along with definitions of the properties they hope to find.
Next, SPARROW collects information about the molecules and their synthesis pathways, then weighs the value of each molecule against the cost of synthesizing a batch of candidates. It automatically selects the best subset of candidates that meet user criteria and finds the most cost-effective synthetic routes for these compounds.
Jenna Fromer, the first author of the paper, said: "It does all these optimizations in one step, so it can capture all these competing goals at the same time."
Multi-functional framework
SPARROW is unique in that it can be integrated Molecular structures designed by humans, existing in virtual catalogs, or never-before-seen molecular structures created by generative AI models.
“We have a variety of different sources of ideas. Part of the appeal of SPARROW is that you can put all these ideas on a level playing field,” Coley added.
Researchers demonstrate SPARROW’s ability to orchestrate molecular design cycles through three case studies. These applications illustrate how SPARROW (1) successfully balances information gain with synthesis costs, (2) captures the nonadditivity of synthesis costs for a batch of molecules, and (3) scales to candidate libraries containing hundreds of molecules.
Illustration: Demonstration of SPARROW’s ability to balance costs and rewards across a library of 14 ASCT2 inhibitor candidates. (Source: Paper)
They found that SPARROW effectively captured the marginal cost of batch synthesis and identified common experimental steps and intermediate chemicals. Furthermore, it can be expanded to handle hundreds of potential molecular candidates.
「In the chemical machine learning community, there are many models that work well for retrosynthesis or molecular property prediction, but how do we actually use them? Our framework aims to leverage the value of these preliminary studies. By creating SPARROW, We hope to guide other researchers in thinking about compound screening using their own cost and utility functions," Fromer said.
In the future, researchers hope to incorporate more complexity into SPARROW. For example, they hope to enable algorithms to take into account that the value of testing a compound may not always be constant. They also want to include more parallel chemical elements in their cost versus value functions.
Reference content: https://news.mit.edu/2024/smarter-way-streamline-drug-discovery-0617
The above is the detailed content of Automatically identify the best molecules and reduce synthesis costs. MIT develops a molecular design decision-making algorithm framework. For more information, please follow other related articles on the PHP Chinese website!

Hot AI Tools

Undresser.AI Undress
AI-powered app for creating realistic nude photos

AI Clothes Remover
Online AI tool for removing clothes from photos.

Undress AI Tool
Undress images for free

Clothoff.io
AI clothes remover

Video Face Swap
Swap faces in any video effortlessly with our completely free AI face swap tool!

Hot Article

Hot Tools

Notepad++7.3.1
Easy-to-use and free code editor

SublimeText3 Chinese version
Chinese version, very easy to use

Zend Studio 13.0.1
Powerful PHP integrated development environment

Dreamweaver CS6
Visual web development tools

SublimeText3 Mac version
God-level code editing software (SublimeText3)

Hot Topics





In modern manufacturing, accurate defect detection is not only the key to ensuring product quality, but also the core of improving production efficiency. However, existing defect detection datasets often lack the accuracy and semantic richness required for practical applications, resulting in models unable to identify specific defect categories or locations. In order to solve this problem, a top research team composed of Hong Kong University of Science and Technology Guangzhou and Simou Technology innovatively developed the "DefectSpectrum" data set, which provides detailed and semantically rich large-scale annotation of industrial defects. As shown in Table 1, compared with other industrial data sets, the "DefectSpectrum" data set provides the most defect annotations (5438 defect samples) and the most detailed defect classification (125 defect categories

The open LLM community is an era when a hundred flowers bloom and compete. You can see Llama-3-70B-Instruct, QWen2-72B-Instruct, Nemotron-4-340B-Instruct, Mixtral-8x22BInstruct-v0.1 and many other excellent performers. Model. However, compared with proprietary large models represented by GPT-4-Turbo, open models still have significant gaps in many fields. In addition to general models, some open models that specialize in key areas have been developed, such as DeepSeek-Coder-V2 for programming and mathematics, and InternVL for visual-language tasks.

Editor |KX To this day, the structural detail and precision determined by crystallography, from simple metals to large membrane proteins, are unmatched by any other method. However, the biggest challenge, the so-called phase problem, remains retrieving phase information from experimentally determined amplitudes. Researchers at the University of Copenhagen in Denmark have developed a deep learning method called PhAI to solve crystal phase problems. A deep learning neural network trained using millions of artificial crystal structures and their corresponding synthetic diffraction data can generate accurate electron density maps. The study shows that this deep learning-based ab initio structural solution method can solve the phase problem at a resolution of only 2 Angstroms, which is equivalent to only 10% to 20% of the data available at atomic resolution, while traditional ab initio Calculation

For AI, Mathematical Olympiad is no longer a problem. On Thursday, Google DeepMind's artificial intelligence completed a feat: using AI to solve the real question of this year's International Mathematical Olympiad IMO, and it was just one step away from winning the gold medal. The IMO competition that just ended last week had six questions involving algebra, combinatorics, geometry and number theory. The hybrid AI system proposed by Google got four questions right and scored 28 points, reaching the silver medal level. Earlier this month, UCLA tenured professor Terence Tao had just promoted the AI Mathematical Olympiad (AIMO Progress Award) with a million-dollar prize. Unexpectedly, the level of AI problem solving had improved to this level before July. Do the questions simultaneously on IMO. The most difficult thing to do correctly is IMO, which has the longest history, the largest scale, and the most negative

Editor | ScienceAI Based on limited clinical data, hundreds of medical algorithms have been approved. Scientists are debating who should test the tools and how best to do so. Devin Singh witnessed a pediatric patient in the emergency room suffer cardiac arrest while waiting for treatment for a long time, which prompted him to explore the application of AI to shorten wait times. Using triage data from SickKids emergency rooms, Singh and colleagues built a series of AI models that provide potential diagnoses and recommend tests. One study showed that these models can speed up doctor visits by 22.3%, speeding up the processing of results by nearly 3 hours per patient requiring a medical test. However, the success of artificial intelligence algorithms in research only verifies this

In 2023, almost every field of AI is evolving at an unprecedented speed. At the same time, AI is constantly pushing the technological boundaries of key tracks such as embodied intelligence and autonomous driving. Under the multi-modal trend, will the situation of Transformer as the mainstream architecture of AI large models be shaken? Why has exploring large models based on MoE (Mixed of Experts) architecture become a new trend in the industry? Can Large Vision Models (LVM) become a new breakthrough in general vision? ...From the 2023 PRO member newsletter of this site released in the past six months, we have selected 10 special interpretations that provide in-depth analysis of technological trends and industrial changes in the above fields to help you achieve your goals in the new year. be prepared. This interpretation comes from Week50 2023

Editor |ScienceAI Question Answering (QA) data set plays a vital role in promoting natural language processing (NLP) research. High-quality QA data sets can not only be used to fine-tune models, but also effectively evaluate the capabilities of large language models (LLM), especially the ability to understand and reason about scientific knowledge. Although there are currently many scientific QA data sets covering medicine, chemistry, biology and other fields, these data sets still have some shortcomings. First, the data form is relatively simple, most of which are multiple-choice questions. They are easy to evaluate, but limit the model's answer selection range and cannot fully test the model's ability to answer scientific questions. In contrast, open-ended Q&A

Editor | KX Retrosynthesis is a critical task in drug discovery and organic synthesis, and AI is increasingly used to speed up the process. Existing AI methods have unsatisfactory performance and limited diversity. In practice, chemical reactions often cause local molecular changes, with considerable overlap between reactants and products. Inspired by this, Hou Tingjun's team at Zhejiang University proposed to redefine single-step retrosynthetic prediction as a molecular string editing task, iteratively refining the target molecular string to generate precursor compounds. And an editing-based retrosynthetic model EditRetro is proposed, which can achieve high-quality and diverse predictions. Extensive experiments show that the model achieves excellent performance on the standard benchmark data set USPTO-50 K, with a top-1 accuracy of 60.8%.
