Datasets
Models
Downloads: 1
Diff of
/Manuscripts/README.md
[000000]
..
[404218]
Switch to side-by-side view
--- a +++ b/Manuscripts/README.md @@ -0,0 +1,343 @@ +<div align="center"> + <p>Autonomous LLM Agent and scalable Reasoning LLM for generating cancer drug industry cost solutions</p> +<div align="center"> + +<div align="center"> + <p>CEO Kevin Kawchak</p> + <p>ChemicalQDevice</p> + <p>San Diego, CA</p> + <p>March 23, 2025</p> + <p>kevink@chemicalqdevice.com</p> +</div> + +<div align="left"> +Every once in a while, a new artificial intelligence technology is released that significantly improves research utility and results. This ’deep research’ tool released by OpenAI in February 2025 is a Large Language Model (LLM) web agent that autonomously queries sites such as PubMed Central (PMC), and generates high quality summaries with verifiable citations. A February 2025 article by Haman M. et al. reported deep research advantages in "analyzing 37 sources-35 of which were found on the PubMed website" using the OpenAI o3 model. Here, ChatGPT 4.5 Deep research summaries regarding five pharmaceutical industry financial categories were found to be 100% in-context with PMC articles, and averaged 1,400 words in 10 minutes with minor issues. Also impressive was the processing of these summaries by the Claude 3.7 Sonnet Extended reasoning model to produce a structured 1,900 word 37 citation report containing detailed economic solutions, which were supported by 6 paragraphs of key insights collaborating multiple author quotations in approx. one minute. In addition, the Claude model produced eleven professional images based on USD or ROI trends, anomalies, and forecasts in three Python scripts. The Claude model possessed an output length that scaled by 3.2x for the report and 6x for code generations vs. the manufacturer’s previous model, was 100% in-context with source data, and included interpretable reasoning summaries. The outputs from ChatGPT 4.5 Deep research served as inputs to 3.7 Sonnet Extended in mitigating model bias amplification that can occur when using results within a single software manufacturer. For transparency, comprehensive generation traceability analyses were conducted for the five summaries, the financial solutions report, and the eleven Python diagrams. + + +<div align="left"> + +<br> + +Kawchak K. Autonomous LLM Agent and scalable Reasoning LLM for generating cancer drug industry cost solutions. Zenodo. 2025; doi:10.5281/zenodo.15072843 + +[](https://doi.org/10.5281/zenodo.15072843) + +--- + + + +<div align="center"> + <p>Cost containment of global monoclonal antibody drugs and cancer clinical trials via LLM focused reasoning</p> +<div align="center"> + +<div align="center"> + <p>CEO Kevin Kawchak</p> + <p>ChemicalQDevice</p> + <p>San Diego, CA</p> + <p>February 25, 2025</p> + <p>kevink@chemicalqdevice.com</p> +</div> + +<div align="left"> +Expenses related to monoclonal antibody drugs worldwide and cancer clinical studies must be reduced in order to increase pharmaceutical industry efficiency. These financial opportunities can be assisted using state-of-the-art Large Language Models (LLMs) for focused report generation and advanced reasoning of solutions and forecasts that are based on authors’ original findings. Here, Claude 3.5 Sonnet utilized 45 articles totaling approximately 357,000 words to effectively generate 45 reports. OpenAI ChatGPT o3-mini processed 15 of the reports to obtain comprehensive monoclonal antibody (mAb) cost solutions and financial forecasts. This included a financial recommendation of mAb biosimilars for a 55.2 percent price per dose decrease vs. a bevacizumab biologic due to Japan financial incentives, as reported by Itoshima H. et al. The 30 additional reports were based on cancer clinical trial cost-effectiveness studies, with ChatGPT o3-mini reasoning to produce tables regarding economic strategies and projections. This included a "Total drug cost avoidance of "$92,662,609" over 10 years" when sponsored clinical trial participation was employed for solid tumors using various mAb therapies, as detailed by Carreras M. et al. 2024. The primary advantages of this comprehensive approach were 1) Efficient report generations by 3.5 Sonnet of nearly 25,000 words in 20 minutes, 2) ChatGPT o3-mini’s linear dependency prompt structure reduced narrative drift with structured outputs in less than 5 minutes, and 3) Ethical AI principles were strengthened: financial data was limited by rigorous prompts, yielding outputs that were highly traceable to source data using either LLM, as opposed to relying on the models’ training data. + +<div align="left"> + +<br> + +Kawchak K. Cost containment of global monoclonal antibody drugs and cancer clinical trials via LLM focused reasoning. Zenodo. 2025; doi:10.5281/zenodo.14968404 This content is a preprint and has not been peer-reviewed. + +[](https://doi.org/10.5281/zenodo.14968404) + +--- + + +<div align="center"> + <p>Gemini Update Clinical decision support based on Bevacizumab cancer trials and pushing the limitations of advanced LLMs</p> +<div align="center"> + +<div align="center"> + <p>CEO Kevin Kawchak</p> + <p>ChemicalQDevice</p> + <p>San Diego, CA</p> + <p>February 2, 2025</p> + <p>kevink@chemicalqdevice.com</p> +</div> + +<div align="left"> +BREAKING: Exact Recall Milestone for Challenging Clinical Decision Support Task. This update to the full January 27, 2025 study significantly improves on Standard 2a results regarding a reasoning Large Language Model’s ability to provide exact recall of author citations and quotations on a comprehensive input [1]. 100 Bevacizumab cancer therapy articles representing over 900K words were previously summarized into 100 reports totaling 49K words, and have now been analyzed by Google Gemini to yield a more accurate clinical decision support (CDS) study. The primarily limitation of the model was less advanced reasoning ability versus current OpenAI o1 and o3-mini models. The relevance of highly accurate quotations and citations is that clinical researchers require consistent and verifiable information along with LLM contextual awareness of clinical studies with associated speed advantages. + +<div align="left"> + +<br> + +Kawchak K. Gemini Update Clinical decision support based on Bevacizumab cancer trials and pushing the limitations of advanced LLMs. Zenodo. 2025; doi:10.5281/zenodo.14968289 This content is a report and has not been peer-reviewed. [Full Study](https://doi.org/10.5281/zenodo.14968162) + +[](https://doi.org/10.5281/zenodo.14968289) + + + +--- + + + + + + +<div align="center"> + <p>Clinical decision support based on Bevacizumab cancer trials and pushing the limitations of advanced LLMs</p> +<div align="center"> + +<div align="center"> + <p>Kevin Kawchak</p> + <p>Chief Executive Officer</p> + <p>ChemicalQDevice</p> + <p>San Diego, CA</p> + <p>January 27, 2025</p> + <p>kevink@chemicalqdevice.com</p> +</div> + +<div align="left"> +An exhaustive study was needed to test the limits of leading Large Language Models (LLMs) using numerous real-world clinical trial outcomes. It was also necessary to provide extensive hallucination studies based on both extracting main points and providing novel AI clinical decision support. Here, 100 Bevacizumab cancer therapy articles representing over 900K words were summarized by the 3.5 Sonnet model into 49K words, which completed a detailed and complex problem of several cancer and study types to press the capabilities of the ChatGPT o1 reasoning model. Report summaries in general followed an effective format, with guardrails to de-identify patient information and numerical data sources attributions to ground the output to the input. The main takeaway was that both LLMs typically remained in-context with the input data when more structured prompts were used, while precise quotations and author name citations were less prominent. These errors were likely due to LLM pressure towards achieving coherence vs. exact recall based on manufacturer inference-time compute settings. Overall, ChatGPT o1 provided state-of-the-art evidence-based Bevacizumab insights regarding clinical efficacy across indications, dosing recommendations, combination therapies, and biomarker-driven selections. + +<div align="left"> + +<br> + +Kawchak K. Clinical decision support based on Bevacizumab cancer trials and pushing the limitations of advanced LLMs. Zenodo. 2025; doi:10.5281/zenodo.14968162 This content is a preprint and has not been peer-reviewed. + +[](https://doi.org/10.5281/zenodo.14968162) + + + + +--- + + + + +<div align="center"> + <p>Cancer vs. Conversational Artificial Intelligence</p> +<div align="center"> + +<div align="center"> + <p>Kevin Kawchak</p> + <p>Chief Executive Officer</p> + <p>ChemicalQDevice</p> + <p>San Diego, CA</p> + <p>December 23, 2024</p> + <p>kevink@chemicalqdevice.com</p> +</div> + +<div align="left"> +Solving cancer mechanisms is challenging due to the complexity of the disease integrated with many approaches that researchers take. In this study, information retrieval was performed on 40 oncological papers to obtain authors' methods regarding the tumor immune microenvironment (TIME) or organ-specific research. 20 TIME summaries were combined and analyzed to yield valuable insights regarding how research based papers compliment information from review papers using Large Language Model (LLM) in-context comparisons, followed by code generation to illustrate each of the authors' methods in a knowledge graph. Next, the 20 combined organ-specific emerging papers impacting historical papers was obtained to serve as a source of data to update a mechanism by Zhang, Y., et al., which was further translated into code by the LLM. The new signaling pathway incorporated four additional authors' area of cancer research followed by the benefit they could have on the original Zhang, Y., et al. pathway. The 40 papers in the study represented over 600,000 words which were focused to specific areas totaling approximately 17,000 words represented by detailed and reproducible reports by Clau-3Opus. ChatGPT o1 provided advanced reasoning based on these authors' methods with extensive correlations and citations. Python or LaTeX code generated by ChatGPT o1 added methods to visualize Conversational AI findings to better understand the intricate nature of cancer research. + +<div align="left"> + +<br> + +Kawchak K. Cancer vs. Conversational Artificial Intelligence. bioRxiv. 2024; doi:10.1101/2024.12.28.630597 This content is a preprint and has not been peer-reviewed. + +[](https://doi.org/10.1101/2024.12.28.630597) + + + + + +--- + + + +<div align="center"> + <p>mAb Bioprocess Engineering In-Context Table Forecasts using Conversational AI Literature Insight Generations</p> +<div align="center"> + +<div align="center"> + <p>Kevin Kawchak</p> + <p>Chief Executive Officer</p> + <p>ChemicalQDevice</p> + <p>San Diego, CA</p> + <p>November 27, 2024</p> + <p>kevink@chemicalqdevice.com</p> +</div> + +<div align="left"> +Bioprocess engineering has incorporated effective AI applications in recent years that consist of traditional approaches to training models on relevant data to then analyze and predict new and unseen data. The missing component has been the ability to process mixed data from an assortment of dissimilar information sources with high contextual awareness to inform the Human-AI team on how LLMs and other authors' methods will further improve performance. Here, real-time web search or document retrieval methods with a max speed multiplier of over 600x by 3.5 Sonnet were obtained vs. the manuscript author regarding monoclonal antibody (mAb) bioprocess engineering kinetics across several models. ChatGPT-4o with an average score of 9/10 was the leader in quality for this task with several detailed reports that were obtained using document search addressing a paper's specific weaknesses being improved with LLMs or two other author's methods. This protocol was applied systematically for each of the other two papers, supported by the other two relevant bioprocess papers. o1-preview's advanced reasoning set a new standard over five other models in processing either 136 extracellular or 101 intracellular metabolite tables, incorporating the analysis of 12 additional paper summaries across two prompts with two table revisions. For extracellular metabolites, o1-preview generated an 18 metabolite table including all metabolite forecasts that were expected to be breakthroughs due to future integration of a LLM or other author's recent methods. The model supported its forecasts with interpretable author citations and quotations for breakthrough metabolites, along with lists of author specific and metabolite specific insights that influenced its conclusions. For intracellular metabolites, o1-preview provided a full 101 metabolite table, matching the number of entries from the original Sukwattananipaat, P., et al. table, including confirmations for each metabolite regarding whether each forecasted value was expected to be a breakthrough. Overall, this work was represented by numerous speed advantages, literature insights to address paper weaknesses, and competent o1-preview in-context table analysis with supporting evidence from leading articles represented by two 9.5/10 scores to lead the first conversational AI mAb bioprocess engineering revolution. + +<div align="left"> + +<br> + +Kawchak K. mAb Bioprocess Engineering In-Context Table Forecasts using Conversational AI Literature Insight Generations. ChemRxiv. 2024; doi:10.26434/chemrxiv-2024-jzbj0 This content is a preprint and has not been peer-reviewed. + +[](https://doi.org/10.26434/chemrxiv-2024-jzbj0) + + + + + +--- + + + + + +<div align="center"> + <p>Monoclonal Antibody Bioprocess Engineering Advancements Using Conversational Artificial Intelligence</p> +<div align="center"> + +<div align="center"> + <p>October 27, 2024</p> + <p>Kevin Kawchak</p> + <p>CEO ChemicalQDevice</p> + <p>kevink@chemicalqdevice.com</p> +</div> + +<div align="left"> +Processing high dimensional and complex monoclonal antibody (mAb) bioprocess data in industry is now more efficient due to conversational AI. The human in the loop approach to Large Language Model (LLM) inferencing with document retrieval and chained outputs is a probable benefit to existing biotechnology workflows. Potential risks of using natural language processing are minimized due to the utility of solving problems with vast amounts of structured and unstructured mixed data that can be verified by the Human-AI team. This novel work demonstrates o1-preview, ChatGPT-4o, L3.1-405B, and 3.5 Sonnet models’ fast and stateof-the-art solutions. In specific, o1-preview provided a response to 16 papers 110x faster than the manuscript author’s time after the number of words were set equal. In addition, ChatGPT-4o was 371x faster than an optimal human researcher to examine and provide an estimate regarding dimension reduction or combinatorial optimization for a recent paper by Kao, M., et al. The third LLM speed advantage of 336x by ChatGPT-4o vs. the manuscript author was achieved using monte carlo simulations and markov chain models performance forecasts and a current paper by Konoike, F., et al. + +Part A featured the individual analysis of 5 recent mAb production papers, which emphasized the proficiency of o1-preview (9.9/10.0), ChatGPT-4o (9.2), and L3.1-405B (9.2) providing a forecast report. Example generations for o1-preview and L3.1-405B typically established connections between using dimension reduction or combinatorial optimization and improving bioprocesses. Part B models generated tables regarding how LLMs can improve numerical data from 5 different papers using monte carlo simulations or markov chain models. An example from ChatGPT-4o (9.0) was substantially more complete, accurate, and convincing than the table provided 3.5 Sonnet (8.0). Part C utilized the report format from Part A combined with the numerical approach from Part B across 6 additional papers, led by o1-preview (9.0) and ChatGPT-4o (8.5). The o1-preview example followed the prompt format well, citing cases of how LLMs will utilize reinforcement learning and bayesian optimization to improve mAb production. The work represents a standard for utilizing a considerable amount of bioprocess data to forecast new results, with the transition into LLMs providing near-real-time production data analysis aided by document retrieval to provide a synergistic effect with existing machine learning techniques. + +<div align="left"> + +<br> + +Kawchak K. Monoclonal Antibody Bioprocess Engineering Advancements Using Conversational Artificial Intelligence. ChemRxiv. 2024; doi:10.26434/chemrxiv-2024-3m7m1 This content is a preprint and has not been peer-reviewed. + +[](https://doi.org/10.26434/chemrxiv-2024-3m7m1) + + + + + +--- + + + + +<div align="center"> + <p>Paclitaxel Biosynthesis AI Breakthrough</p> +<div align="center"> + +<div align="center"> + <p>October 3, 2024</p> + <p>Kevin Kawchak</p> + <p>CEO ChemicalQDevice</p> + <p>kevink@chemicalqdevice.com</p> +</div> + +<div align="left"> +Paclitaxel, C<sub>47</sub>H<sub>51</sub>NO<sub>14</sub>, biosynthesis is an active area of research due to ongoing progress towards more sustainable and environmentally friendly production of the drug compound. Recent literature details the characterization of enzymes that play a role in synthesis, optimization of growth media, and RNA related regulatory mechanisms. The method of PhD students spending excessive time performing literature reviews to discover new findings is obsolete due to faster and high quality state of the art conversational AI. In this study, approximate AI times were obtained regarding how long would it take the fastest human researcher to read, analyze, extract information, and type a high quality 250 word answer; with the fastest time of 1,380 seconds being used as a standard reference. The slowest AI generation in the study was 79.19s by ChatGPT-4o, which was still over 17x faster than the optimal human performance time. Here, a paclitaxel biosynthesis breakthrough was illustrated twice using LLMs and LMMs. In the first instance, full length papers were summarized by AI models – with the finding that AI provided more detailed answers across entire papers, generating over 10x longer descriptions and 12x faster times compared to the manuscript author’s methods to summarize abstracts. + +The outputs of individual AI generated answers yielded a 10 Paper Summary with 6,322 words, and served as the input for eight separate prompts, which provided valuable insight regarding both emerging and historical views of paclitaxel retrobiosynthesis, engineering microorganisms, as well as top 10 new research recommendations, and top 10 challenges for this area. The second paclitaxel biosynthesis advancement was demonstrated with a speed of 752 seconds for 36 generations compared to the single optimal human response of 1,380 seconds. Top models received an average AI judge score of 9.5 by ChatGPT-4o for Part A; a score of 9.3 by o1-preview, L3.1-405B, and ChatGPT-4o for Part B; and a score of 9.3 by ChatGPT-4o and 3.5 Sonnet, followed by a score of 9.2 for Wiz8x22B for Part C. These superior results have primarily been afforded by OpenAI, Claude.ai, and Meta AI new model releases in late 2024 that have helped to advance the paclitaxel biosynthesis field. The presence of speedups with more detailed answers over optimal human responses is supported by advanced cloud hardware that processes high dimensional and complex data continually to solve combinatorial problems such as those in this study using 15 different prompts across 163 generations. + +<div align="left"> + +<br> + +Kawchak K. Paclitaxel Biosynthesis AI Breakthrough. ChemRxiv. 2024; doi:10.26434/chemrxiv-2024-pqjd3 This content is a preprint and has not been peer-reviewed. Creative Commons Attribution 4.0 International. + +[](https://doi.org/10.26434/chemrxiv-2024-pqjd3) + + + + +--- + +<div align="center"> + <p>High Dimensional and Complex Spectrometric Data Analysis of an Organic Compound using Large Multimodal Models and Chained Outputs</p> +<div align="center"> + +<div align="center"> + <p>September 12, 2024</p> + <p>Kevin Kawchak</p> + <p>CEO ChemicalQDevice</p> + <p>kevink@chemicalqdevice.com</p> +</div> + +<div align="left"> +Large Multimodal Models (LMMs) possess the ability to analyze chemical spectra of an organic compound using state of the art conversational AI. These outputs can then be chained together and introduced as a text input for other LLMs or LMMs to predict the compound name. Here, a challenging 15 carbon molecule problem with 13 complex and high dimensional chemical spectra were analyzed as images by unmodified versions of Claude 3.5 Sonnet and OpenAI ChatGPT-4o models. ScholarGPT judged the responses across the 13 spectra with an average score of 9.01/10, and the highest response scores per individual spectra for 3.5 Sonnet or GPT-4o were used as the text-based chain. For Part B, the chain was then combined with two different prompt formats and the molecular formula to 8 different LMMs or LLMs which produced new compound predictions. 3.5 Sonnet had the highest proficiency in utilizing the formula simultaneously with complex data for three identical compound generations across two prompts, but was likely limited by the quality regarding the chain of 13, primarily with data from 6 2D NMR Spectra. 3.5 Sonnet's compound prediction was then further improved in Part C by utilizing manual chained explanations of the spectra by the author to yield what is believed to be the correct structure with stereochemistry to the unknown problem. To the author's best knowledge, this is the first LMM to generate the C15H22O2 drug compound derivative (S)-ibuprofen ethylester using high dimensional data from 13 detailed spectra. The purpose of this study was to utilize cutting edge natural language processing techniques to evaluate an advanced chemical structure consisting of IR, 1H-NMR, 13C-NMR, DEPT-NMR, GCOSY60, GTOCSY, GHMQC, GHMBC, GNOESY, and expanded views of spectra. + +<div align="left"> + +<br> + +Kawchak K. High Dimensional and Complex Spectrometric Data Analysis of an Organic Compound using Large Multimodal Models and Chained Outputs. ChemRxiv. 2024; doi:10.26434/chemrxiv-2024-06gf1 This content is a preprint and has not been peer-reviewed. + +[](https://doi.org/10.26434/chemrxiv-2024-06gf1) + + + + + + + +--- + + + + +<div align="center"> + <p>LMM Spectrometric Determination of an Organic Compound</p> +<div align="center"> + +<div align="center"> + <p>August 26, 2024</p> + <p>Kevin Kawchak</p> + <p>CEO ChemicalQDevice</p> + <p>kevink@chemicalqdevice.com</p> +</div> + +<div align="left"> +Many machine learning models used in academia and industry that identify organic compounds typically lack the ability to converse over prompts and results, and also require expertise across a number of steps to obtain answers. The purpose of this study was primarily to gain insight into the advantages of current unmodified state of the art Large Multimodal Models (LMMs) across several prompts containing multiple spectra of varying difficulty to evaluate the impact of training data, reasoning, and speed. These readily available and easy to use software for the identification of an organic compound based on a molecular formula and spectra were found to be reproducible across three similar LMMs. To the author's best knowledge, this marks the first time that three GPT variants were each able to correctly identify the organic compound quinoline using a variety of different spectroscopic images. The results were obtained using a 2-step process consisting of a) Uploading high resolution spectral images, and b) Submitting a text prompt with the images that requested a compound determination. The main findings were that 1) Four LMMs provided rationale step-by-step interpretations of 1H-NMR, 13C-NMR, and 3 DEPT-NMR spectra from Prompt A, 2) Three of these LMMs, led by a GPT-5 preview model, combined these interpretations into the correct chemical structure with Prompt A, and 3) Two of these LMMs achieved a top score of 5/5 for also generating sequential explanations reflecting the order of the provided spectra along with most of the correct spectral and molecular formula explanations. + +<div align="left"> + +<br> + +Kawchak K. LMM Spectrometric Determination of an Organic Compound. ChemRxiv. 2024; doi:10.26434/chemrxiv-2024-qtnkj This content is a preprint and has not been peer-reviewed. + +[](https://doi.org/10.26434/chemrxiv-2024-qtnkj) + + + + + + + +--- + + +<div align="center"> + <p>LMM Chemical Research with Document Retrieval</p> +<div align="center"> + +<div align="center"> + <p>Kevin Kawchak</p> + <p>Chief Executive Officer</p> + <p>ChemicalQDevice</p> + <p>San Diego, CA</p> + <p>August 12, 2024</p> + <p>kevink@chemicalqdevice.com</p> +</div> + +<div align="left"> +Chemical research is more effectively progressed using Large Multimodal Models (LMMs) combined with Document Retrieval and recently published literature. The methods described here illustrate significant strides over previously tested Large Language Model (LLM) multi-document workflows for characterization assistance and generating new reactions. Here, 3.5 Sonnet, ScholarGPT, and ChatGPT 4o LMMs processed either 5 images or 5 supplementary documents from leading 2024 journals. Each of the three models performed inference on a detailed prompt to produce a response that included context from attachments. In addition, the LMMs were not provided with which of the 5 files contained the answer. The main findings were that 3.5 Sonnet had an average score of 9.8 for images, while two judges awarded high scores to ChatGPT 4o (9.7, 9.4) and ScholarGPT (9.5, 9.4) for document analysis. Judging was performed by a human evaluator for the image uploads, with document processing evaluated by Llama 3.1 405B and Nemotron 4 340B LLMs which correlated well and improved explainability. Highlights include 3.5 Sonnet's ability to interpret a Two-dimensional Nuclear Magnetic Resonance (2D NMR) spectrum accurately, along with Judge Llama 3.1's ability to provide consistent formatted scores with explanations. The results shown here help illustrate AI's continued revitalization of the established chemical research field. + +<div align="left"> + +<br> + +Kawchak K. LMM Chemical Research with Document Retrieval. ChemRxiv. 2024; doi:10.26434/chemrxiv-2024-p91gm This content is a preprint and has not been peer-reviewed. + +[](https://doi.org/10.26434/chemrxiv-2024-p91gm) + + +--- + + + +## AI Applications for Drug Industry +LLM and LLM agent pharmaceutical industry applications + +[](https://doi.org/10.5281/zenodo.13273141)