The Useful Information for Biological Resources (UIBR^®) aims to discover useful biological resources by integrating and standardizing information on their various uses. UIBR^® collects useful information for biological resources scattered in books, old literatures, reports, and published papers. It integrates terminologies, such as local names, scientific names, drug names of biological resources, and disease names, with the standardized species list provided by ISDB™.

AI platform for Useful Information for Biological Resources (UIBR-AI^®) is the platform for predicting novel usages of plant resources with the aid of machine learning and deep learning algorithms based on the massive and diverse data stored in the UIBR^® as well as related databases, including ISDB™.

ISDB™ is an integrated system for managing species information originated from diverse species databases containing scientific names and synonyms. Collected data were integrated into the standardized species list as a standardized species data structure.

Infoboss Reference Database (RefDB™) is an integrated database for managing and analyzing publications originated from various publication databases, such as PubMed. It contains i) various crawlers to collect references periodically, ii) the pipeline to build non-redundant reference information as well as iii) the web-based environment to choose and to analyze proper publications interested. In addition, RefDB™ provides the standardized reference information for many databases and systems of Infoboss.

Patent Analysis Platform (PAP™) is an integrated platform for patents. Several crawlers for collecting patents efficiently have been implemented in this system. Additional analysis of patents was conducted together with RefDB™.

MetaPool™ is the platform for classifying publications and patents originated from RefDB™ and PAP™, respectively, with the aids of machine learning and deep learning algorithms.

MetaEve-AI™ was designed for evaluating economical values of plant resources in various aspects with the aid of the machine learning engine. It will provide the guide of many unused plant resources which have potential to be commercialized.

Genome Information System (GeIS™) is the optimized system containing the four main pipelines for analyzing whole genome sequences efficiently. Various analysis tools implemented in the four pipelines can be executed through the GeIS™ web interface and the analysis results can be accessed and reanalyzed via the GlobalScrap^®.

GenomeArchive^® is a standardized genome database which covering whole genome and organelle genome sequences. GenomeArchive^® covers all organism's genome sequences from virus to human. GenomeArchive^® also supports the multiple versions of genome sequences by providing previous version of sequences. All genetic elements, such as gene, transcript, and proteins, were accessible via genome browser. Modular design allows GenomeArchive^® to provide these databases and interfaces to various system for supporting establishment of systems utilizing genome sequences cost effectively. Currently, 14,198 whole genomes and 3,180 organelle genomes were collected in GenomeArchive^®.

SSRDB™ was designed for identifying and analyzing simple sequence repeats from the whole and organelle genome sequences managed by GenomeArchive^®. It contains the identification pipeline and various web-based analysis tools of SSRs. In addition, modular design of SSRDB™ provides the way to present identified SSRs in the different systems efficiently.

Gene Family Database™ is a collection of independent gene family databases, such as GATA-TFDB, P450DB, and OSC-DB. Gene Family Database provides a standardized pipeline protocol to identify specific gene family with various filtering conditions and analysis pipelines including sequence alignment and construction of phylogenetic tree. In addition, basic format of web interfaces will provide the easy access to identified gene family members. Moreover, GlobalScrap^® will provide the versatile environment to analyze them in detail to understand their characteristics. Two publications (R1 and R2) have been published based on this pipeline till now.

Infoboss Pathway Database (IPDB™) was constructed for managing scattered biochemical reaction data with the non-redundant compound names and enzyme classifications. The pipeline named as Reaction Maker constructs the global reaction maps as well as compound-specific pathways. These generated pathways are utilized for predicting secondary metabolites from the whole genome sequences together with MetaPre-AI^®. Additional biochemical pathways are also integrated into the IPDB™ for uncovering biochemical pathways of all phytocompounds.

The Phytocompound Database (PhytoCompDB™) aims for standardizing phytocompounds with proper species, parts, and conditions with precise references. It can provide global view of phytocompounds based on the accumulated researches so far. We will try to utilize these identified phytocompounds in various aspects, including combining pathway information as well as whole genome data. We will also try to integrate the databases which has alrady established with specific purposes for expanding the coverage of phytocompound data.

MetaPre-AI^® is the system to predict phytocompounds from the plant whole genome sequences with ensemble engine containing traditional identification pipeline as well as multiple pipelines developed based on machine learning and deep learning.

Platform for EC Number Prediction Tools (PECT™) is an integrated platform for analyzing enzyme sequences. Currently, 31,120,928 enzyme sequences were collected from several major database including BRENDA. Several classifiers for predicting enzyme functions as EC number were implemented to evaluate their accuracy for genes predicted from newly sequenced genome sequences. Moreover, annotation pipeline of EC number was also implemented to annotate and to feed as a training dataset of the classifier. Web interface and GlobalScrap^® provide the analysis environment on the web.

MetaTarget^® was designed for predicting mode of actions as well as their biological activities of phytocompounds. In the case of prediction of mode of actions, our engine was constructed with consideration of mixtures, which is plant extract. For training these engines, published papers, molecular mechanisms including signal transduction pathway, related gene sequences, phytocompounds, as well as protein-small molecule interactions were utilized.

Meta-ISM^® was designed for confirming various phytocompounds from plant extracts with small number of HPLC-based experiments. It will be used for confirming phytocompounds which were predicted from the whole genome sequences by MetaPre-AI^®.

MetaEx™ was designed for confirming biological functions of plant extracts or phytocompounds especially predicted by MetaTarget^®. It consists of various cell-based experiments, so called as in vitro experiments, and in vivo experiments. Standardized experiments for each function, such as anti-inflammatory, are used for providing the training dataset of MetaTarget^® for precise prediction of its biological functionalities.