ChatPandaGPT integrates a Large Language Model (LLM) and chat functionality to provide information and answer questions related to molecular biology, therapeutic target discovery, and pharmaceutical development. With a chatbot-like interface, the LLM allows it to give more personalized and relevant responses to each user based on specific needs and interests. ChatPandaGPT is aware of general biological data and has been augmented with specific information from the PandaOmics Knowledge Graph, ensuring accuracy in each response.

The user can interact with ChatPandaGPT by selecting one of the predefined questions like "What are the potential risks and benefits of targeting gene X?". Alternatively, users can formulate their own queries and receive relevant answers. ChatPandaGPT can be found on the Publications tab of the following pages: Gene, Disease, Gene-Disease Association. Please refer to the ChatPandaGPT manual section for more detailed information.

Drug target assessment must include diverse considerations related to biology, chemistry, technology and business. While there is no single set of questions that are comprehensive for every context, there are a few that very commonly arise. Is there a strong, mechanistic link between the target and the disease? Is it feasible to design and manufacture a drug against this target? What properties would a drug against this target need to have to be competitive in the market, and would it make sense in our company's portfolio?

PandaOmics 4.0 introduces a groundbreaking and revolutionary approach to creating Gene-Disease reports in the context of drug target investigation. By combining omics data from an internal database with the power of Large Language Models (LLMs), PandaOmics now enables users to automatically create a detailed report encompassing all available pieces of evidence for gene-disease associations. This new approach plays a pivotal role in the PandaOmics workflow, providing critical step when users are on the quest for their targets and diseases of interest.

In version 4.0 of PandaOmics, users can quantitatively prioritize diseases for a given target with a variety of configurable multi-omics-based and text-based scores. Top-ranked diseases can be viewed as candidates in the context of indication expansion for existing drugs or identifying novel indications for promising targets. Indications can be categorized based on either pathology groups (such as oncology or infection) or by organ systems (such as the respiratory or digestive system) for convenience. Indication prioritization functionality can be found on the Indication Prioritization tab of any Gene page.

The latest release of PandaOmics brings a wealth of genetics-related enhancements to our platform. In response to the ever-growing importance of genetics in drug discovery and biomedical research, we've introduced a dedicated Genetics tab, providing a streamlined interface for in-depth genetic exploration. This tab allows users to access gene- and variant-level information from several genetic databases and can be found on any Disease page.

Other enchantments include allowing users to visualize genes carrying relevant genetic variants on the Knowledge Graphs. We also improved genetics-related scores for the ranking of potential therapeutic targets in oncology and other therapeutic areas. These advancements aim to significantly enhance genetic research capabilities within PandaOmics, facilitating more informed decision-making and a deeper understanding of the role of genetics in disease mechanisms.

In PandaOmics 3.0 we introduced the Knowledge Graph, a result of automated analyses of scientific publications, which allows users to visualize the relationships between genes, diseases, chemical compounds, and biological processes. By delving into Knowledge Graph, researchers can gain insights into the molecular underpinnings of diseases and the known roles of genes and compounds in the disease context. This resource assists in making informed decisions about prospective drug targets and biomarkers.

In this release, we bring significant enhancements to the Knowledge Graph. Our database has been updated to include the latest publications available up to Sep 2023. The Knowledge Graph now offers the capability to visually identify whether a gene harbors genetic variants known to be linked to a particular disease. Additionally, we have improved the user interface by introducing new designs for the legend and graph layout selector. The Knowledge Graph is accessible on the Publications tab of the Meta-Analysis, Gene, Disease, and Gene-Disease Association pages.

PandaOmics 4.0 introduces the capability for users to upload custom, gene-level methylomics datasets. These datasets can undergo processing through a standard PandaOmics data analysis pipeline, thus greatly enhancing support for multi-omics data analysis. The inclusion of the custom methylome upload feature seamlessly integrates user-proprietary data into PandaOmics Target ID results. PandaOmics accommodates the upload of both M-values and B-values, and it automatically converts B-values to M-values when necessary. Additionally, during the regularly scheduled updates for the PandaOmics platform, the methylation microarray processing for public datasets has been expanded to support both Illumina 850k and 935k platforms. New datasets based on these platforms are now collected from GEO monthly in addition to the 450k datasets supported in earlier versions. These updates significantly enrich the methylomics analyses within the platform.

In PandaOmics 4.0, we have incorporated over 300 manually curated disease-specific Meta-Analyses, with a particular focus on cancer, rare and genetic diseases, as well as inflammatory and infectious diseases. Pre-calculated Target ID scores and Compound ID scores are available on their respective tabs within each Disease page. We ensure the regular updates of the pre-calculated Disease pages on a weekly basis.

Over 2,000 high-quality omics datasets covering 71,000 samples have been added to the platform and are readily available for target discovery analyses. In this release, we have placed a special focus on methylation data support. More than 500 datasets for three platforms (450k, 850k, and 935k) were added to the database. Also, several proteomics datasets across 40+ diseases have been added, covering more than 2,000 samples. Overall this improvement has led to a ten-fold increase in the number of pre-calculated multi-omics disease projects available to the user compared to PandaOmics 3.0.

This release introduces significant improvements in search capabilities across the user interface. Specifically, we've enhanced pathway search functionality by enabling searches across various meta-data fields, including pathway name, description, keywords, and pathway content such as gene names. In addition, when querying for genes within the Target ID or Expression Analysis sections, users will receive results irrespective of the current filtering criteria. The Publications tab supports a comprehensive search approach, encompassing publication titles, abstracts, author surnames, and keywords. Furthermore, we've removed the constraint that previously limited the display to the top 500 publications associated with Genes or Diseases, offering a more extensive view of relevant publications

The druggability filters (Small molecules, Antibodies, Safety, and Novelty) on both the Meta-Analysis and Disease pages can now be customized to accommodate any combination of values. Furthermore, we've enhanced the functionality of Meta-Analysis, allowing users to add or reattach comparisons and modify the therapeutic area for previously conducted Meta-analyses, creating an editable copy for greater flexibility. Additionally, navigation within the Pharma.AI platform is now more seamless, with direct links to Chemistry42 and InClinico conveniently located in the upper left corner of every PandaOmics screen. Lastly, the Expression Analysis page for each Meta-Analysis now includes the option to export differential analysis results in a comma-separated text file, providing greater utility for your research needs.

The neural network of special architecture, the so-called transformer extracts the relevant information from the list of scientific publications and produces a nice-looking graph that visually links genes, diseases, chemical compounds, and biological processes. The knowledge graph is accessible on the Publications tab of the Gene, Disease, and Gene-Disease Association pages. Explore the knowledge graph to gain insights into the molecular mechanisms of the disease, and biological activities of genes and compounds in the context of the disease, and decide on the most promising drug targets and biomarkers. Click on genes and diseases, rearrange nodes, and use zoom and pan functions to come up with publication-ready images. Hovering the mouse over the graph's edge brings up tooltips with the paper citations supporting the corresponding relation in the graph.

While users always have the option to analyze datasets independently from each other and then combine the expression and pathway results in a Meta-analysis (macro-aggregation), the cross-dataset comparison feature allows merging sample groups together across multiple datasets on earlier steps of analysis (micro-aggregation). This function is available on the Dataset tab in the Data Manager. Batch correction can also be enabled to improve the quality of the results and correct for possible batch effects. The user is also provided with quality metrics and PCA plot, informing whether the batch correction has been performed successfully. Even datasets without paired normal control samples can now be used for the subsequent analysis.

Several tabs on the Disease and Meta-analysis pages have been significantly updated. Access the sample number and omics data types available in the Meta-Analysis directly from the Meta-analysis summary tab. Availability of microarray, RNA-seq, proteomics, methylation, and GWAS data can be found in the omics data summary card. This also applies to the curated disease reports available through the main search. Use filters similar to the Target ID tab to narrow down the list of genes in the Expression Analysis tab. For example, one can quickly obtain a list of kinases with resolved 3D protein structures showing differential expression across various datasets. The publication-ready expression heatmap can now be exported as an image in vector or raster graphics format.

In PandaOmics 3.0 we have significantly improved our named-entity recognition technology. This results in more accurate automatic detection of various biomedical entities like genes, diseases, drugs, etc. Our web crawlers regularly scan the scientific literature, clinical trials, patents, and grant applications to provide you with the most up-to-date information about promising targets and therapeutics. This also ensures AI-based target prioritization scores operate at their maximum performance with the addition of frequent updates to text and clinical trial data.

For this release, our AI experts have compiled and analyzed -omics datasets for most if not all the diseases included in the pipelines of the top 20 pharma companies, globally. This means that PandaOmics users are provided with pre-calculated, out-of-the-box reports for promising diseases and conditions. PandaOmics 3.0 gives immediate access to your prioritized lists of targets and compounds without the need to perform the data analysis yourself. In case of interest in a specific therapeutic condition, the user can rely on the comprehensive PandaOmics database to build new projects. In Version 3.0, more than 1000+ high-quality datasets have been added to the PandaOmics library.

Pre-calculated Target ID and Compound ID are now available for all the diseases. They can be accessed through the Disease page. Just open any disease page and go to the corresponding tabs in the left sidebar or use the links from the Disease Summary tab. The pre-calculated analysis is based on the text data, GWAS, and biological graphs. Manually curated, high-quality Gene expression datasets are also used for omics scores calculation for more than 200 diseases.

Manually curated datasets checkmark is introduced in the Dataset tab on the Disease pages. The datasets flagged with check marks have passed manual quality control of meta-data annotation, study design relevance to the disease, and data quality. Save time on inspecting the datasets manually and select the ones with check marks if available to ensure high quality of the results.

Omics dataset summary tab on the Disease and Meta-analysis pages provides the statistics of the omics data used for subsequent steps of analysis including Target ID and Compound ID.
The total number of genes in the dataset and the portion of the differentially expressed genes is shown. It is also possible to group Comparisons, e.g. by the tissue of origin or other properties of the source samples and datasets. The Comparison groups can be further used to aggregate the results in Target ID.

Target ID: Switching on/off Comparisons is now allowed in the Target ID interface without reassembling and recalculating the entire Meta-analysis. Just use the new feature of the Target ID dashboard to turn on and off Comparisons or Comparison groups, and get the updated ranking immediately. This feature is specifically useful to compare the target ID outputs calculated using public data only versus the proprietary datasets uploaded by the user. Another example is switching between the Target ID results for different tissues.

2 additional AI scores in Target ID were added in the omics section of the target discovery dashboard. Heterogeneous Graph Walk (HeroWalk) is a guided random walk-based approach that is applied to a heterogeneous gene-disease graph. Matrix Factorization is a collaborative filtering algorithm widely used in recommender systems applied to a gene-disease interaction graph. The new scores strengthen the AI component and result in a better performance of the target hypothesis generation engine while preserving the general layout and backward compatibility of the Target ID tool.

More features are added in the Target ID section of the application. Presets allow saving custom states of the filtering options (druggability, tissue-specific expression, list of scores used for ranking, etc.) and applying them in one click to any disease. A collection of default presets e.g. targets for repurposing, and novel targets, is available. A gene of interest can be found using the Gene search box. The definition of the filtering criteria in a druggability traffic light widget is reworked.