Sure! Here’s the translation of your text into American English:
—
In the field of pharmaceutical science, the development of new medications is a complex and lengthy process. However, companies like Genentech and AstraZeneca are implementing artificial intelligence (AI) and generative tools to streamline this task. By utilizing the Amazon Bedrock platform, these organizations are establishing specific workflows that cover everything from early identification of drug targets to interaction with healthcare providers.
To address more complex use cases, the use of Strands Agents SDK is being considered, an open-source platform that employs a model-based approach for the development and operation of AI agents. Strands Agents is compatible with various model providers, allowing for integration that includes specific language accesses and internal use. This approach facilitates the implementation of agents in environments hosting Python applications.
Recently, a new research assistant has been developed that integrates Strands Agents and Amazon Bedrock. This assistant can simultaneously query multiple scientific databases through the Model Context Protocol (MCP). Additionally, it can synthesize findings and generate detailed reports on pharmacological targets, disease mechanisms, and therapeutic areas. This resource is available as an example within the open-source health and life sciences agent tool.
The solution connects high-performance models with relevant life sciences data from recognized sources such as arXiv, PubMed, and ChEMBL. Research indicates that small teams of AI agents can outperform a single, broader agent. The architecture of this solution is based on an orchestrating agent that manages user queries and redirects them to specialized sub-agents, depending on the need to retrieve information or generate plans and syntheses.
Currently, the assistant is in the testing phase and features a chat interface that allows for the evaluation of its effectiveness in various research inquiries. For example, a user might request a report on the HER2 receptor in the context of breast cancer research. The assistant is capable of generating a work plan that includes searching for recent news, scientific articles, and ongoing clinical trials, in addition to synthesizing all this information into a single report.
In addition to report generation, this assistant eliminates the need to define a strict process for each task, as it has the ability to determine which tools to use and in what sequence. With the exponential growth of scientific information, tools like Strands Agents are becoming indispensable resources in the search for new drugs.
—
Let me know if you need any further assistance!
Referrer: MiMub in Spanish
