Patent Information Users Group, Inc.

Highlights of the Program of the  American Chemical Society Fall 2020 National Meeting (August 17-20, 2020)

Notes: Access to Meeting Platform with the final technical program referenced  below requires login with ACS  [website] ID;  a preliminary technical program is accessible without registration. The full CINF symposia program & abstracts (with contacts of speakers [in bold] ) is published in Chemical Information BULLETIN, 2020 v.72 No. 3. pp. 31-71, and available here.

 

Division of Chemistry and the Law (CHAL)

August 18, 2020, 11:30 AM - 12:00 PM    Broadcast

Changing landscape of patent eligible subject matter: Where do we go from here?

Emily Gabranski (Finnegan, Henderson, Farabow, Garrett & Dunner, LLP, Boston, MA)

The past decade has seen a significant and unpredictable shift in the scope of patentable subject matter. With the looming possibility of further guidance from the Supreme Court, legislative changes from Congress, and new guidance from the USPTO, the law remains uncertain and unpredictable. In this presentation we discuss the cases that control today and try to chart the path forward for chemical and biotech innovators.

https://www.abstractsonline.com/pp8/index.html#!/9308/presentation/5682

Related webinar: Finnegan. The Current State of Patentable Subject Matter in the United States. Recorded Mar. 4, 2020 (1:02:56)

 August 19, 2020, 1:03 PM - 1:23 PM Broadcast

Antibody patenting strategies

Michael Spellberg Spellberg (Lathrop GPM LLP, Boston, Massachusetts)

This lecture provides an overview of patent strategies relating to therapeutic antibodies, with a particular focus on antibody-drug conjugates (ADCs). Antibodies are a rapidly expanding area medicine that offer highly targeted therapies to a wide range of diseases, including cancer, autoimmune disorders, and infectious disease. This presentation will be useful to those with all levels of experience in pharmaceutical patent law.

https://www.abstractsonline.com/pp8/index.html#!/9308/presentation/6859

Related webinars:

Kilburn & Strode/World Intellectual Property Review Antibody Patents: Maximising Value in Europe and Beyond; (Recorded Aug 5 2020)

Mathys & Squire LLP. Antibodies and Non-Obviousness at the EPO;  (Recorded 7 July 2020)

BSKB/World Intellectual Property Review: Protecting the Pot of Gold: Patent Applications for Therapeutic Antibodies (Recorded Jan 30 2018)

 

Division of Chemical Information (CINF)

Symposium: Machine Learning & Artificial Intelligence in Computational Chemistry

August 17, 2020, 10:03 AM - 10:23 AM       Broadcast

Synthetic feasibility and de novo molecular generation and optimization

Wenhao Gao, Connor Coley  (Chemical Engineering, Massachusetts Institute of Technology, Cambridge, MA)

There is substantial interest in de novo molecular generation and optimization techniques as a way to propose new molecules during early-stage drug discovery. Deep generative models and other inverse design techniques allow for the multi-objective optimization of surrogate models (e.g., for activity, druglikeness) without relying on brute-force screening of a virtual chemical library. However, their utility is limited by an ignorance of synthesizability. This talk will describe an evaluation of several state-of-the-art methods in terms of their abilities to generate synthesizable molecules as judged by a data-driven chemical synthesis planning program. I will summarize emerging methods that explicitly integrate synthetic feasibility into their generative processes, which is an important step toward overcoming this limitation.

https://www.abstractsonline.com/pp8/index.html#!/9308/presentation/5428

Gao, W., Coley, C.W., 2020. The Synthesizability of Molecules Proposed by Generative Models. J. Chem. Inf. Model. https://doi.org/10.1021/acs.jcim.0c00174

Methods: ASKCOS.ASKCOS is an open-source software framework that integrates efforts to generalize known chemistry to new substrates by learning to apply retrosynthetic transformations, to identify suitable reaction conditions, and to evaluate whether reactions are likely to be successful when attempted experimentally. Data-driven models within ASKCOS are trained on millions of reactions from the U.S. Patent and Trademark Office (USPTO) and Reaxys databases. The core retrosynthetic capabilities rely on the recursive application of algorithmically extracted reaction templates encoded as SMARTS patterns. … Importantly, ASKCOS has both programmatic and graphical interfaces to enable thousands of compounds to be processed without human intervention.

ASKCOS Synthesis Predictor [Forward Prediction] Module

Predict the outcome of a chemical reaction by specifying the reactants and context. This module allows you to use the Reaxys-trained template-based or USPTO-trained template-free forward predictor to anticipate the outcomes of an arbitrary chemical reaction…(Tutorial)

 Cited databases:

Lowe, D. (2017). Chemical reactions from US patents (1976-Sep2016). Figshare. https://doi.org/10.6084/M9.FIGSHARE.5104873.v1

Reactions extracted by text-mining from United States patents published between 1976 and September 2016. The reactions are available as CML [Chemical Markup Language] or reaction SMILES. Note that the reactions SMILES are derived from the CML. 

Ruddigkeit, L., van Deursen, R., Blum, L.C., Reymond, J.-L., 2012. Enumeration of 166 Billion Organic Small Molecules in the Chemical Universe Database GDB-17. J. Chem. Inf. Model. 52, 2864–2875. https://doi.org/10.1021/ci300415d

GDB-17 is an open database containing 166.4 billion enumerated molecules with up to 17 heavy atoms of C, N, O,S, and halogens.

 Coley, C.W., Thomas, D.A., Lummiss, J.A.M., Jaworski, J.N., Breen, C.P., Schultz, V., Hart, T., Fishman, J.S., Rogers, L., Gao, H., Hicklin, R.W., Plehiers, P.P., Byington, J., Piotti, J.S., Green, W.H., Hart, A.J., Jamison, T.F., Jensen, K.F., 2019. A robotic platform for flow synthesis of organic compounds informed by AI planning. Science 365. https://doi.org/10.1126/science.aax1566

We developed an open source software suite for CASP trained on millions of reactions from the Reaxys database and the U.S. Patent and Trademark Office. The software was designed to generalize known chemical reactions to new substrates by learning to apply retrosynthetic transformations, to identify suitable reaction conditions, and to evaluate whether reactions are likely to be successful when attempted experimentally. Suggested routes partially populate CRFs, which require additional details from chemist users to define residence times, stoichiometries, and concentrations that are compatible with continuous flow. To execute these syntheses, a robotic arm assembles modular process units (reactors and separators) into a continuous flow path according to the desired process configuration defined in the CRF. The robot also connects reagent lines and computer-controlled pumps to reactor inlets through a fluidic switchboard. When that is completed, the system primes the lines and starts the synthesis. After a specified synthesis time, the system flushes the lines with a cleaning solvent, and the robotic arm disconnects reagent lines and removes process modules to their appropriate storage locations.

 

August 17, 2020, 12:21 AM - 12:41 AM              On Demand Oral

ChEMU shared task: chemical entity recognition and event extraction of chemical reactions from patents

Christian Druckenbrodt, Saber Akhondi, Camilo Thorne (Elsevier, Frankfurt, Germany), Karin Verspoor, Dat Nguyen, Zenan Zhai (University of Melbourne, Melbourne, New South Wales, Australia)

We introduce a new chemical information extraction shared task, named ChEMU [Cheminformatics Elsevier Melbourne University], part of the 11th Conference and Labs of the Evaluation Forum (CLEF-2020). ChEMU proposes two key information extraction tasks over chemical reactions from patents. Task 1 --- named entity recognition (NER) --- is to identify specific types of chemical compounds, i.e. to assign the label of a chemical compound according to the role for which the chemical compound plays within a chemical reaction. Task 2 --- event extraction over chemical reactions (EE) --- involves on the other hand event trigger detection and argument recognition. We will publicly release reaction-specific gold standards --- that we describe in this pressentation --- derived from patent literature and annotated by chemists (for NER and EE, resp.) in early 2020. Thereafter academic or industrial teams working in the field will be encouraged to participate in a shared evaluation campaign, by developing and contributing NER and EE models. The models will be then evaluated (by measuring their recall, precision and F1 scores), compared and jointly discussed at CLEF-2020.

https://www.abstractsonline.com/pp8/index.html#!/9308/presentation/1890

 

Related publications

Dat Quoc Nguyen, Zenan Zhai, Hiyori Yoshikawa, Biaoyan Fang, Christian Druckenbrodt, Camilo Thorne, Ralph Hoessel, Saber A. Akhondi, Trevor Cohn, Timothy Baldwin and Karin Verspoor. ChEMU: Named Entity Recognition and Event Extraction of Chemical Reactions from Patents. Presentation at ECIR2020 [European Conference on Information Retrieval 2020], video (https://www.youtube.com/watch?v=eHJTkFUxJgg&t=9944s at 2:45:44)

Proceedings:  Nguyen, D. Q.; Zhai, Z.; Yoshikawa, H.; Fang, B.; Druckenbrodt, C.; Thorne, C.; Hoessel, R.; Akhondi, S. A.; Cohn, T.; Baldwin, T.; Verspoor, K. ChEMU: Named Entity Recognition and Event Extraction of Chemical Reactions from Patents. In Advances in Information Retrieval; Jose, J. M., Yilmaz, E., Magalhães, J., Castells, P., Ferro, N., Silva, M. J., Martins, F., Eds.; Lecture Notes in Computer Science; Springer International Publishing: Cham, 2020; pp 572–579. https://doi.org/10.1007/978-3-030-45442-5_74. (Submitted copy)

ChEMU: Cheminformatics Elsevier Melbourne University. Information Extraction from Chemical Patent. A project website (https://chemu-patent-ie.github.io/https://chemu-patent-ie.github.io/)

Task 1: Named Entity Recognitioninvolves identifying chemical compounds as well as their types in context, i.e., to assign the label of a chemical compound according to the role which the compound plays within a chemical reaction.

Task 2: Event extraction over chemical reactions involves event trigger detection and argument recognition

Background publications:

1. Zhai Z, Nguyen DQ, Akhondi S, Thorne C, Druckenbrodt C, Cohn T, Gregory M and Verspoor K. (2019) Improving Chemical Named Entity Recognition in Patents with Contextualized Word Embeddings. Proceedings of the Workshop on Biomedical Natural Language Processing (BioNLP) at ACL 2019. https://www.aclweb.org/anthology/W19-5035.pdf

2. Yoshikawa H, Verspoor K, Baldwin T, Nguyen DQ, Zhai Z, Zkhondi S, Thorne C, Druckenbrodt C. (2019) Detecting Chemical Reaction Schemes in Patents. Australian Language Technology Association Workshop (ALTA 2019). Sydney, Australia, December 2019. https://www.aclweb.org/anthology/U19-1014.pdf

Evaluation Lab meeting @CLEF 2020, Thessaloniki, Greece: September 22-25 2020 [online only event)

 

Symposium: Reaction Prediction & Synthesis Planning

August 17, 2020, 12:01 AM - 12:21 AM On Demand Oral

Comprehensive search for compounds and chemical reactions in big query

Stephen Boyer¹, Ian Wetherbee², Timo Böhme³, Matthias Irmer³, Christoph Ruttkies³, Lutz Weber⁴
1. Collabra Inc., San Jose, California, United States. 2. Google, Mountain View, California, United States. 3. OntoChem GmbH, Halle/Saale, Germany. 4. IT, OntoChem, Germering, Germany

Google BigQuery (BQ) provides a platform for a growing amount of scientific information in various open access data collections. To take advantage of the data in such a large and diverse aggregation, we have enabled structure- and ontology-based searches. The SciWalker Open Data project maintains BQ chemistry-related tables of compounds (currently numbering 131 million) that are updated daily from mentions found in the text or images of patents. These tables include compound SMILES, InChIs, and InChI-Keys as well as unique public ontology concept identifiers, OCIDs. Similarly, we register chemical reactions from patents into a table containing reaction SMILES, RInChIs, the short RInChI-Keys, as well as OCIDs. Both compounds and reactions are classified into respective ontology classes available in BQ for searching.
The main purpose of these tables is to link information from numerous BigQuery tables via standard SQL queries. For example, we can ask ontology-based questions like “which sesquiterpenes have been used in clinical trials as treatments?”
We have also used the Chemistry Development Kit (CDK) to create fingerprints for the compounds and reactions, implementing structure and substructure search capability on top of BigQuery.

See another presenataions of these authors below.

Related publications:

Boyer, S. (IBM Almaden Research Center) Leveraging IP data for its scientific content. [Computer curation of patents & the scientific literature in the Digital Age]. Presentation at the 11^th Luxembourg Day of Intellectual Property, Apr. 26, 2018, 88 p. https://ipil.lu/wp-content/uploads/2018/04/3_Boyer_ipday2018_Luxemburg.pdf

OntoChem IT Solutions. SkiWalker: Integrating Open Access and Private Sources, 2019.Presentation on IC-SDV 2019 April 8, 2019, 14 p.   https://www.slideshare.net/Haxel/icsdv-2019-ontochem

 OntoChem IT Solutions, 2019. Open access WebAPI registration and retrieval of ontology concepts. Presentation at InChI Trust 2019, August 23, 2019. 17 p. https://www.inchi-trust.org/wp/wp-content/uploads/2019/12/4.-OntoChem-InChI_Trust_2019-08-23-v2-1.pdf

 Lutz Weber (OntoChem GmbH). SciWalker: Semantic Chemistry Search. Video (5:32)  Presented June 4, 2020 at ChemAxon Cheminfo Stories 2020 Partner Session [133 mln compouns from ~200 mln docs, incl. 123 mln. DOCDB abstracts and 24 mln WO, US, EP full-text patents, video at  1:20 & 1:27, searchable at ] https://www.sciwalker.com]

 OntoChem’s SkiWalker Open Data website (https://www.sciwalker.com) (powered by Chemaxon) (Documentation) (a first public release Aug.6, 2019)(“A web search engine that implements advanced information retrieval and extraction from abstracts, full text articles, patents, and web pages. It uses a set of multihierarchical dictionaries for annotation and ontological concept searching. It can be used to annotate any public or internal repository of heterogeneous documents. The search interface allows queries with logical combinations of free text and ontological terms”. Quoted from: Wendy Warr's Report: Cheminfo Stories 2020 (July 2020))

Antunes, L., Khorsandian, F., Krabbe, E., Sampson, S., Wetherbee, I., 2018. Big Data & Public Databases for Patent Research and Analysis. 2 p. https://ipo.org/wp-content/uploads/2018/09/Patent-Searching-Public-Databases.pdf ( Demonstrating how Google BigQuery “can be used in answering queries about patent information”)

August 19, 2020, 10:03 AM - 10:23 AM  Broadcast

Predicting reaction sequences: Deep neural networks and reaxys data bases

Friedrich Kroll  (Elsevier, Frankfurt, Deutschland (DEU), Germany)

 …Reaxys is a software tool that contains well curated FAIR (Findable, Accessible, Interoperable, Reusable) scientific information and teaches chemists how to find and apply the right literature for chemical transformations, identify analytical information and most importantly educates, how to develop retrosynthesis schemes swiftly. The possibilities to evaluate and compare different routes, yields and conditions to molecules are exceptional and allow educators to illustrate retrosynthesis in organic chemistry classes on a day to day basis. Furthermore, computer-aided retrosynthesis tools, which can predict reactions to novel molecules correctly, are part of Reaxys. The ‘deep learning’, neuronal networks produce schemes of sequences of reactions to the desired compounds. This novel AI tool has processed nearly every reaction ever published (> 15 million) and has the potential to transform the way organic chemists work in the future. In an assessment, various synthesis routes generated were tested in a double-blinded trial with 45 organic chemists from two institutes in China and Germany and the routes have proven to be scientifically sound and robust. This unique approach to retrosynthesis will lead to an increase of the success rate in synthetic organic chemistry and should have an enormous benefit in terms of discovering sustainable chemical solutions and minimizing expenditure.

https://www.abstractsonline.com/pp8/index.html#!/9308/presentation/5944

 

Related:

Reaxys Predictive Retrosynthesis: Redesigning the approach to synthetic chemistry (Product Summary)

Based on:

Segler, M.H.S., Preuss, M., Waller, M.P., 2018. Planning chemical syntheses with deep neural networks and symbolic AI. Nature 555, 604–610. https://doi.org/10.1038/nature25978

To plan the syntheses of small organic molecules, chemists use retrosynthesis, a problem-solving technique in which target molecules are recursively transformed into increasingly simpler precursors. Computer-aided retrosynthesis would be a valuable tool but at present it is slow and provides results of unsatisfactory quality. Here we use Monte Carlo tree search and symbolic artificial intelligence (AI) to discover retrosynthetic routes. We combined Monte Carlo tree search with an expansion policy network that guides the search, and a filter network to pre-select the most promising retrosynthetic steps. These deep neural networks were trained on essentially all reactions ever published in organic chemistry. Our system solves for almost twice as many molecules, thirty times faster than the traditional computer-aided search method, which is based on extracted rules and hand-designed heuristics. In a double-blind AB test, chemists on average considered our computer-generated routes to be equivalent to reported literature routes.

Elsevier. Reaxis Fact Sheet. 2019. 6 p. [>118M  substances; >49 M reactions; >59 M docs from 16,000 chemistry related journals;  1.5 M patents (WO, US,EU,JP, KR, CN, TW)]

 

August 19, 2020, 11:00 AM - 11:20 AM Broadcast

Reaction Transformers for Fingerprints, Classification and Atom-Mapping

Philippe Schwaller, Daniel Probst, Ben Hoover, Alain Vaucher, Vishnu Nair, David Kreutter, Hendrik Strobelt, Teodoro Laino, Jean-Louis Reymond ( IBM Research -- Zurich, Rueschlikon, Switzerlan; University of Bern, Bern, Switzerland; IBM Research -- Cambridge / MIT-IBM Lab, Cambridge, MA)

Self-supervised language models called transformers have recently revolutionized natural language processing and show tremendous potential when applied to text-based representations of chemical reactions. The patterns in chemical reactions are learned by predicting masked parts of reaction SMILES. The pretrained models can then be specialized on a task like reaction classification, where they reach unprecedented accuracies. Not only can specific outputs of the transformer models serve as fingerprints to map the chemical reaction space without the need of knowing the reaction center or distinguishing between reactants and reagents, but they can also be used to recover the rearrangement between reactant and product atoms. By opening the black-box using detailed visual analysis, we discovered that the transformer models learned atom-mapping without supervision. Atom-mapping, known to be an NP-hard problem, is necessary for making chemical reaction data better machine-accessible and crucial for graph- and template-based reaction prediction and synthesis planning approaches. Here, we present an attention-guided reaction mapper that shows remarkable performance in terms of speed and accuracy, even for strongly imbalanced reactions as typically found in patents. This work is the first demonstration of knowledge extraction from a self-supervised language model with a direct practical application in the chemical reaction domain.

https://www.abstractsonline.com/pp8/index.html#!/9308/presentation/5947

 

Related article:

Schwaller, P., Petraglia, R., Zullo, V., Nair, V.H., Haeuselmann, R.A., Pisoni, R., Bekas, C., Iuliano, A., Laino, T., 2020. Predicting retrosynthetic pathways using transformer-based models and a hyper-graph exploration strategy. Chem. Sci. 11, 3316–3325. https://doi.org/10.1039/C9SC05704H

August 19, 2020, 11:40 AM - 12:00 PM  Broadcast

SAVI a la carte: Moving toward molecules on demand by AI. The development of the SLICE (Smarts and Logic In ChEmistry) language

Victorien Delannee, Marc Nicklaus(National Cancer Institute, Frederick, Maryland)

The current version of SAVI (Synthetically Accessible Virtual Inventory) is an expert system for the generation of very large libraries of easily synthesizable molecules, utilizing the LHASA project's language pair CHMTRN/PATRAN for synthetic knowledge rules. One unique strength of CHMTRN is the use of a logic validated by chemists to predict both possible failure and success of reactions. However, CHMTRN is an old unstructured and non-standardized language working retrosynthetically, which presents challenges in a forward-synthetic context. To overcome these limitations, we have developed SLICE (Smarts and Logic In ChEmistry), which combines SMARTS with a logic language. SMARTS describes the molecular patterns, while the logic allows reasoning by defining rules such as "IF statement THEN action." This new language is highly inter-operable, suitable in both a forward and retro-synthetic context, easily readable and extended SMARTS capabilities. We show that this language can be used in the context of SAVI to speed up the first-step reaction prediction by making it forward-synthetic and can also be used in a retro-synthetic context to retrieve the synthetic road predicted by AI.

https://www.abstractsonline.com/pp8/index.html#!/9308/presentation/5949

Related Resources:

Patel, H., Ihlenfeldt, W., Judson, P., Moroz, Y.S., Pevzner, Y., Peach, M., Tarasova, N., Nicklaus, M., 2020. Synthetically Accessible Virtual Inventory (SAVI). https://doi.org/10.26434/chemrxiv.12185559.v1

Synthetically Accessible Virtual Inventory (SAVI) Database SAVI-2020 Full File Series - April 2020

1.75 billion proposed products with reactions generated in the first full enumeration of the SAVI project

This version utilizes: (a) a set of transforms with rich chemical context annotation including functional group reactivity data and scoring systems (based on the LHASA project, programmed in CHMTRN/PATRAN; see https://doi.org/10.26434/chemrxiv.11439984.v1); (b) a set of readily available building blocks in gram quantities (Enamine, Kyiv, Ukraine); (c) the chemoinformatics toolkit CACTVS with custom development (Xemistry GmbH, Glashütten, Germany). For generation of the SAVI-2020 set of products, 53 transforms were used, applied to approx. 150,000 building blocks in single-step reactions. The CHMTRN/PATRAN source code of the transforms can be downloaded here. For a more detailed description of the SAVI project, see https://doi.org/10.26434/chemrxiv.12185559.v1.

https://cactus.nci.nih.gov/download/savi_download/

 

August 19, 2020, 1:40 PM - 2:00 PM             Broadcast

Computer-aided synthesis planning & ASKCOS

Connor Coley (Chemical Engineering, Massachusetts Institute of Technology, Cambridge, MA)

 Machine learning and artificial intelligence have enabled new data-driven approaches to CASP [Computer Aided SynthesisPlanning] where statistical models are trained directly on published experimental data. A group of researchers at MIT developed several of these tools in an integrated software suite, ASKCOS, that is capable of proposing retrosynthetic routes to new molecules, proposing reaction conditions for each step, and assessing the likelihood of experimental success. This talk will summarize the development of ASKCOS, algorithmic contributions, user adoption, user experiences, and outstanding challenges. I will emphasize a new template-free, graph-based approach to proposing retrosynthetic disconnections.

https://www.abstractsonline.com/pp8/index.html#!/9308/presentation/6085

 See additional info listed to related presentation above.

August 19, 2020, 2:00 PM - 2:20 PM         Broadcast

Overcoming conflicts and dilemmas in computer-aided synthesis design

Orr Ravitz, Gary Gustafson, Paul Peters (CAS, Columbus, Ohio)

 Computer-aided synthesis design (CASD) tools are now routinely used by synthetic chemists across different fields and industries. In SciFinder-n alone, thousands of chemists are using predictive retrosynthesis on a daily basis as a means of boosting creativity, and as a way to expose and thoroughly and efficiently explore alternative strategies and methods for making new and known molecules. …In this talk, we will share observations and insights from over a year of usage of retrosynthesis in SciFinder-n… We will provide an overview of the ongoing research and development programs at CAS. In stereo-selectivity and regio-selectivity we will discuss machine-learning vs statistical approaches, as well as the benefits and limitations of deriving the knowledge from the entire corpus vs learning the selectivity for individual reactions. The relationship between solution-set size and diversity and the search algorithm, as well as the dependency on scoring and distance functions will be demonstrated. Finally we will discuss the impact of the data-set on the scope and accuracy of solutions, and the criticality of evidence on the user-experience.

https://www.abstractsonline.com/pp8/index.html#!/9308/presentation/6086

Related publications:

CAS. SciFinder-n Retrosynthesis Webinar (Demonstration segment, 33:32). Recorded July 1, 2020.

Peters, P. (CAS) Making sense of predicted routes: the use of data as evidence for predictions in SciFinder. Presentation on AI for Reaction Outcome & Synthetic Route Prediction 2020 Conference (AI React 2020)‎ March 9-11, 2020. 15 p.

CAS, 2020  Retrosynthetic Analysis and Synthesis Planning in SciFinderⁿ  [“For new or known molecules, SciFinderⁿ will perform a full retrosynthetic analysis utilizing the … CAS collection of reactions….; includes a video Scifinder-n: Synthesis Planning  (01:37)]

CAS. CAS Rolls Out New Predictive Retrosynthetic Capabilities in SciFinderⁿ. (Press Release Jan. 13, 2020)

This computer-aided synthetic design (CASD) solution utilizes AI technology, powered by CAS’s …collection of scientist-curated reaction content and leverages John Wiley and Sons, Inc.’s … ChemPlanner technology to now identify predicted retrosynthetic routes for both known and novel compounds.

 See also: Warr, W., 2017. What do synthetic chemists want from their reaction systems? Chemical Information BULLETIN v. 69. No.4 (with summaries of Orr Ravitz and Jonathan Taylor (both CAS) on SciFinderⁿ and ChemPlanner integration; Chemplanner has been referenced by:

Law, J.; Zsoldos, Z.; Simon, A.; Reid, D.; Liu, Y.; Khew, S. Y.; Johnson, A. P.; Major, S.; Wade, R. A.; Ando, H. Y. Route Designer: A Retrosynthetic Analysis Tool Utilizing Automated Retrosynthetic Rule Generation. J. Chem. Inf. Model. 2009, 49 (3), 593–602. https://doi.org/10.1021/ci800228y. [Semantic Scholar]

Cook, A.; Johnson, A. P.; Law, J.; Mirzazadeh, M.; Ravitz, O.; Simon, A. Computer-Aided Synthesis Design: 40 Years On. WIREs Computational Molecular Science 2012, 2 (1), 79–107. https://doi.org/10.1002/wcms.61.

 

Symposium: Moving Chemistry from the Lab into the Open

August 20, 2020, 10:40 AM - 11:00 AM              Broadcast

Integration of chemistry with everything else

Ian Wetherbee (Google), Lutz Weber (OntoChem, Germering, Germany), Evan Bolton (NCBI), Steve Walker(Googlae), Stephen Boyer, Jane Frommer (Collabra Inc., San Jose, CA)

The discovery and association of molecules and their attributes are important for strengthening scientific awareness in commercial and societal settings. With increasing demand for molecular information as input for machine learning - leading to yet further discovery - the identification, organization and availability of the world's molecular content are all in demand. Together Google Patents and OntoChem are making a significant contribution of computer-curated molecular data to NIH PubChem, upholding the FAIR data principles of Findable, Accessible, Interoperable, and Reusable. This donation to NIH will make available for the first time machine-curated data derived from text and images of previously uncurated patents from around the world. Additionally, Google is providing a BigQuery platform for the integration of molecular data with content from other worldwide resources. Areas to benefit from this platform include medicine, economics, agriculture, climate change, pharma, chemistry and the law.

https://www.abstractsonline.com/pp8/index.html#!/9308/presentation/6208

See another presenataions of these authors above.

Related publications:

Boyer, S. (IBM Almaden Research Center) Leveraging IP data for its scientific content. [Computer curation of patents & the scientific literature in the Digital Age]. Presentation at the 11^th Luxembourg Day of Intellectual Property, Apr. 26, 2018, 88 p. https://ipil.lu/wp-content/uploads/2018/04/3_Boyer_ipday2018_Luxemburg.pdf

Southan, Christopher. Structure searching in Google patents Bio <-> Chem Blog. (posted 27th August 2019)

Southan, Christopher. IBM chemistry in PubChem. Bio <-> Chem Blog. (posted 18th October 2018; updated May 2019)

WIPO Contributes Millions of Searchable Chemical Formulas to Database at U.S. National Institutes of Health. Patentscope News,  March 25, 2020 [Pubchem News March 16, 2020] [> 16 mln structures from Patenscope where chemical exact and sub-structure searches have been enabled  in collaboration with InfoChem, a DeepMatter Group company]

Update: An abstract of Steve Boyer's presentation "Comprehensive search for compounds and chemical reactions in big query" has been added. (8/20/2020 9:12 AM)