Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating a user interface presentation using a life sciences research platform. In one aspect, a method includes: receiving input data for multiple life sciences experiments within a research domain, where input data includes a collection of experimental settings for the life sciences experiments, automatically generating multiple machine learning models for the research domain, where each machine learning model predicts a value for an experimental outcome metric, automatically selecting a final machine learning model within the research domain based on performance measures, and generating a user interface presentation of explainability data that explains a contribution of each setting in a collection of settings to the value for the experimental outcome metric predicted by the final machine learning model.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for maintaining data of a hierarchical representation of a protein complex that includes components and automatically generating missing component data of the hierarchical representation of the protein complex. By storing format templates that define structural constraints and hierarchical relationships between components of a protein complex, the described techniques enable standardized and modular representations that reduce data redundancy and improve consistency across protein designs. By automatically generating missing component data using biological transformations, the system improves data integrity, reduces manual errors, and supports scalable protein design and production.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for automatic or semi-automatic end-to-end data analysis of bioprocess output data. In particular, using a bioinformatics platform subsystem in conjunction with a laboratory computer subsystem and an electronic laboratory notebook subsystem for automatic analysis of bioprocess output data within a unified platform system. The platform system receives user input that includes a request to perform an automatic analysis process and ultimately updates an electronic laboratory notebook entry that includes output data generated by the analysis. Because the platform system is capable of in real-time automatic handling of a plurality of datasets types from a plurality of bioprocesses, the system can simplify and accelerate bioprocess output data analysis and can be used in feedback-driven optimization of bioprocess procedures.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for automatic or semi-automatic bioprocess design, execution, and analysis. In one aspect, a system comprises receiving a request to execute a sequence of unit operations specified by a recipe representing a bioprocess, wherein each unit operation defines one or more steps to be performed, generating, for each step of each unit operation, a worksheet representing laboratory effects of performing the step and having a worksheet interface for displaying bioprocess data and receiving execution data, including initializing, for each worksheet, a corresponding electronic laboratory notebook object in an underlying electronic laboratory notebook subsystem, receiving execution data entered into the worksheet interface for a particular step of a first unit operation, and updating, in the electronic laboratory notebook subsystem, the corresponding electronic laboratory notebook object using the execution data.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for a computing platform that supports bioprocess design, execution, analysis, and technology transfer. One of the methods includes receiving, by a bioinformatics platform from a production facility, a facilities specification that represents properties and capabilities of each item of equipment of a plurality of items of equipment at the production facility. A selection of a recipe to be transferred to the production facility as a production plan is received. A matching process is performed between unit operations of the selected recipe and equipment represented by the facilities specification to convert each equipment-agnostic unit operation of the recipe into an equipment-specific production task to be performed by a particular item of equipment in the production facility. A production plan is generated having a plurality of equipment-specific production tasks for the production facility.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for automatic or semi-automatic bioprocess design, execution, and analysis. In one aspect, a system comprises receiving a request to execute a sequence of unit operations specified by a recipe representing a bioprocess, wherein each unit operation defines one or more steps to be performed, generating, for each step of each unit operation, a worksheet representing laboratory effects of performing the step and having a worksheet interface for displaying bioprocess data and receiving execution data, including initializing, for each worksheet, a corresponding electronic laboratory notebook object in an underlying electronic laboratory notebook subsystem, receiving execution data entered into the worksheet interface for a particular step of a first unit operation, and updating, in the electronic laboratory notebook subsystem, the corresponding electronic laboratory notebook object using the execution data.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for searching chemical structures. One of the methods includes receiving a fingerprint map for each chemical structure of a collection of chemical structures, generating an inverted index for the collection of chemical structures, receiving a request to search the inverted index for matching chemical structures, obtaining a query fingerprint for the query structure, computing a minimum number of bits and a maximum number of bits that must be present in respective fingerprints of candidate structures in the collection of chemical structures in order for a candidate structure to be considered a match for the query structure, and traversing the inverted index to identify candidate structures having fingerprints with a number of bits in between the minimum number of bits and the maximum number of bits.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for updating a molecular database. In one aspect, a method comprises: receiving a request to update a target molecule component stored in the molecular database; updating the data defining a target molecule represented by the target molecule component; identifying a set of target component links that reference the target molecule component; identifying, for each of the target component links, a respective validation criterion associated with the target component link; determining, for each of the target component links, a new validation status for the target component link by evaluating the validation criterion associated with the target component link; and updating the current validation status of each target component link to the new validation status for the target component link.