Patent Information Users Group, Inc.

Part 1 of Highlights of 2021 Fall ACS Meeting includes 7 selected presentations  CHAL Symposiums Aug. 22-23, 2021 and selected presentation of the following CINF Symposiums for Aug. 22, 2021: Chemistry Babylon (6 presentations),  Reaction Informatics (5) and ML/AI for Organic Chemistry (1)

See Part 2 of Highlights of 2021 Fall ACS Meeting which includes selected presentations of CINF symposia for Aug. 23-24, 2021 and posted in the reply to this PIUG-PF post.

[CHAL] Division of Chemistry & the Law   (Program)

Aug. 22, 2021

Strengthening Your Patent Rights in Light of Recent Federal Circuit Court Decisions Symposium [CHAL]

04:30pm - 05:00pm ET
Update on Federal Circuit and PTAB case law affecting chemical scientists
Dr Andrew Berks, Ph.D., J.D. (Gallet Dreyer Berkey LLP, New York, NY)
…This talk will review recent Federal Circuit Court decisions on patents and provide insights to scientists on how to write and review patent applications such that the patents that are granted from these applications would be successful in a court of law.
Note: Andy Berks was PIUG Chair in 1992-1994


06:00pm - 06:30pm ET
How Recent Federal Court Decisions and USPTO Proceedings are Impacting Intellectual Property Rights in the Chemical Arts Fields
Katherine Rubino, Caldwell Intellectual Property, Boston, MA
… Discussed in this presentation will be an overview of various recent cases that have been decided in Federal Courts, and the impact of these cases on intellectual property rights…specifically … in the chemical arts space.

Aug. 23, 2021

08:00am - 10:00am ET
Post issuance proceedings: Certificates of Correction, Supplemental Examination, and Reissue Patents [CHAL]
Jonathan Kennedy, McKee, Voorhees & Sease PLC, Des Moines, Iowa, US [Presenter]
…This symposium will cover common types of issues identified by patentees with their issued patents and options available to address them including, Certificates of Correction, Supplemental Examination, and Reissue Patents 

Hot Topics in IP Law Symposium [CHAL]

02:00pm - 02:30pm ET
Understanding inventorship
Anthony Zucchero, Meunir Carlin & Curfman, Atlanta, GA
… This presentation will outline the legal standard for inventorship using several key Federal Circuit Cases, with a particular focus on cases related to the chemical and pharmaceutical arts.

02:30pm - 03:00pm ET
Is your structure sound? Testing the limits of the structural obviousness framework
Dr. Kimberlynn B. Davis, J.D., Ph.D., Kilpatrick Townsend & Stockton LLP
To determine whether structurally “close” compounds are obvious for patentability purposes, the Federal Circuit established a framework that has been consistently applied since its inception, often with favorable results for the patentee in Hatch-Waxman litigations. Patent applicants are also able to utilize this framework to establish the patentability of novel compounds, though examiners continue to refute the applicability of the framework by pointing to older standards. This presentation will provide an overview of the framework and its application in key cases…

03:00pm - 03:30pm
Is the BPCIA the new ANDA?
Shannon Patrick, Finnegan Henderson Farabow Garrett and Dunner LLP, Washington, DC
ANDA litigation has become almost a more specialized subset of patent law that focuses on the filings of new drug applications and the statutorily infringing abbreviated new drug applications. Recently, Congress has enacted the BPCIA which creates a similar scheme to the laws regulating ANDA litigations for biological products. This presentation will discuss this new abbreviated pathway for biosimilars and try to elaborate on the steps to the BPCIA patent dance.

04:30pm - 05:00pm ET
The importance of the priority application
Jonas Jarvholm, Ballard Spahr LLP, Atlanta, GA
…This talk will focus on the disclosure in the priority application compared to the disclosure in a conversion application and how this relates to prior art and intervening inventions.
https://acs.digitellinc.com/acs/live/21/page/411/1?eventSearchInput=&eventSearchTrack[]=63


[CINF] Division of Chemical Information  (Program)
August, 22, 2021
Chemistry Babylon: Overcoming Language Barriers to Share Chemical Information Symposium [w/ abstracts]
Organizers: Susan K Cardinal, David Deng
In this symposium we will explore the roots of technical language translation, discuss services that support chemists whose primary language is not English and find out how close we are to a "universal translator."


04:40pm - 05:05pm ET
Semantic and chemistry structure search on 250 million scientific documents
Lutz Weber, OntoChem, Halle (Saale), Sachsen-Anhalt, Germany
… In the current presentation we will demonstrate semantic and chemistry structure search in more than 110 million scientific articles and 130 million patents. Based on our context sensitive named entity recognition, we also extract specific relationships between those indexed entities - such as chemical reactions or biological activities of chemical compounds. These relationships are stored and searched using a novel, patented database technology - xfactDB - that allows to efficiently store complex and structured scientific data.
Examples  provided: which natural products from edible plants support the treatment of diabetes ; which non-lactame natural product compound classes are likely to exhibit anti-infective properties; what is the likelihood of a succesful clinical development of a given chemical compound ?

05:05pm - 05:30pm ET
IUPAC Apocalypse: How changing standards can make chemical nomenclature knowledge and software obsolete
Roger Sayle, NextMove Software Ltd, Cambridge, Cambridgeshire, UK
The latest 2013 revision of IUPAC's "blue book" includes one obscure recommendation rule  (sec. P-52.2.5.1, [which] requires the use of Phane Nomenclature for organic molecules for which it is (sufficiently) applicable, …including names of many drug-like molecules. For example, WHO’ INN for  Danuglipron is (now) officially named (1²S)-9⁴-cyano-9²-fluoro-7-oxa-3(1,2)-benzimidazola-6(2,6)-pyridina-5(1,4)-piperidina-1(2)-oxetana-9(1)-benzenanonaphane-3⁶-carboxylic acid.

05:35pm - 06:00pm
Extracting chemicals, reactions and other named entities from non-English documents
Richard Gowers, John Mayfield; Roger Sayle, NextMove Software Ltd, Cambridge, UK
...In this talk we will present the latest advances in NextMove Software’s Leadmine software, and how it can be used to extract chemical data from sources in a variety of languages including Chinese, Japanese and Korean. We will discuss the approaches used to process chemical nomenclature and reaction descriptions, the algorithms used for correcting perceived errors, and how these are affected when considering the interlingual dimension.


06:00pm - 06:25pm
Challenges of chemical name recognition in Japanese documents
Bela Pukanszky, Árpád Figyelmesi, Chemaxon, Budapest, Hungary
…We investigate the bidirectional conversion of chemical and biochemical structures and names in various formats [in Japanese documents which use the combination of katakana, kanji, roman characters and numbers], in addition to IUPAC and traditional name support, also recognizing common names, and CAS registry numbers. We enable indexing documents, facilitating retrieval of the exact location of the chemical data extracted for each document.


07:10pm - 07:35pm ET
Roots of finding chemical translations: Tools that uncover technical reports
Chris Sherratt, MIT, Cambridge, MA
… This presentation will briefly review the resources used to find English translations in the technical report literature [which decades ago expanded access to chemistry's international literature]; the tools' digital availability through the HathiTrust Emergency Access Temporary Service (ETAS); and how one particular resource might be used to uncover an institution’s own historical, chemical research.


07:35pm - 08:00pm
Mitigating challenges in multilingual communication for the scientific community: Using machine translation to search and understand the world of scientific information
Ian Wetherbee (Google Inc, Mountain View, CA); Stephen Boyer (Collabra Inc, San Jose, CA); Jane Frommer (IBM Almaden Research Center, San Jose, CA)
…Google Patents implements Google Translate to bring the world’s patent and scientific literature to the public. This talk presents technical advances in architecture and quality of neural machine translation of international scientific documents, from searching to understanding content.
https://acs.digitellinc.com/acs/live/21/page/411/1?eventSearchInput=&eventSearchTrack[]=83&eventSearchTag[]=487&eventSearchRoom[]=190


August 22, 2021
02:00pm - 04:00pm ET
Reactions: Representation, Planning and Robotics Symposium [CINF] [w/ abstracts]
[w/ abstracts]Organizers: Gerd Blanke, Prof. Jonathan M Goodman, Guenter Grethe, Christina Keil
What information is necessary to record about a reaction? How can machine learning and AI support the process? … Topics include… RInChI, UDM, AI methods for analysing reaction databases, data exchange formats to run and extract data from synthesis, and robots for synthesis execution.
https://acs.digitellinc.com/acs/live/21/page/411/1?eventSearchInput=&eventSearchTrack[]=83&eventSearchRoom[]=85

11:25am - 11:50am
Extracting chemical reaction information from patent documents
 Lutz Weber, Matthias Irmer; Shadrack Barnabas; Claudia Bobach; Christoph Ruttkies, OntoChem (Germany)
… The BQ SciWalker Open Data project [1] maintains BQ chemistry-related tables of … (>4 million chemical reactions that are updated daily from mentions found in the text or images of patents. In the present talk we present our efforts to extract reaction SMILES, RInChIs, and RInChI-Keys as well as unique public Ontology Concept IDentifiers, OCIDs [2] for each unique chemical reaction. Extracted reactions are classified into their respective reaction ontology classes using compregensive SMIRKS representations of reaction transforms like SMARTS in case of compounds [3]. A particular challenge of such reaction classification is a correct mapping of the atoms in reacting centers. We have compared the precision of several methods for such atom mappings.
[1] https://console.cloud.google.com/bigquery?project=sciwalker-open-data
[2] https://registry.identifiers.org/registry/ocid
[3] Automated compound classification using a chemical ontology. J. Cheminform. 2012, 4(1), 40.

11:50am - 12:15pm
RInChI: Collecting and integrating more reaction data
Prof.  Jonathan M Goodman, University of Cambridge (UK); Gerd Blanke, StructurePendium Technoligies GmbH, Germany
… In this talk we will report the results of an investigation into how effectively the RInChI [The Reaction InChI] is able to link different reports of the same reaction and distinguish similar but different reactions. We will also discuss how it is possible to gather more data for databases and what additional information could usefully be associated with the standard RInChI. Finally, we will discuss how developments in the InChI itself (stereochemistry, organometallics and variable structures) might strengthen the RInChI's ability to identify reactions.


02:00pm - 02:25pm ET
Inferring missing molecules in incomplete chemical equations
Dr. Alain C. Vaucher, Presenter; Philippe Schwaller; Alessandra Toniato; Teodoro Laino  (IBM Research Europe, Rüschlikon, Zürich, Switzerland)
…We present a deep-learning model based on the transformer architecture that infers the molecules in partial reaction SMILES strings [5]. This model does not contain any chemical knowledge except the one learned from the data during training. We illustrate its application for data curation, as well as its use for forward and retrosynthesis prediction.
[5] Vaucher, A.C., Schwaller, P., Laino, T., 2020. Completion of Partial Reaction Equations . https://doi.org/10.26434/chemrxiv.13273310.v1


02:25pm - 02:50pm
Deep learning prediction of reaction conditions
Matthew Clark¹, Eric Gilbert¹; Frederik van den Broek²
(1)Elsevier Inc Philadelphia, PA; (2) Elsevier BV, Amsterdam, Noord-Holland, Netherlands
… The Reaxys database has reactions and conditions reported for over 20 million organic reactions reported over the last 200 years. In many cases the same reaction has been reported using diverse conditions by many different authors. We have used these diverse conditions to train a neural-network that predicts the reagents, catalysts, solvents, time, temperature to carry out any proposed reaction, as well as yield for the products. We will present our results and share our experience from building these models.


03:15pm - 03:40pm ET
The unified data model [UDM]
Gerd Blanke (StructurePendium Technoligies GmbH, Germany); Elena Herzog & Markus Fischer (Elsevier BV, Amsterdam, Netherlands); Jaroslaw Tomczak (Informatics Unlimited Ltd, Cambridge, UK)
…The Unified Data Model (UDM) aims at improving data exchange and integration of chemical reaction data for efficient search, visualization, analysis and predictive modeling…. The UDM project has been fully moved to open source under the MIT license this year.
In our presentation, we will discuss the current status of the data exchange and integration needs and how the UDM can facilitate an easy exchange of data in a variety of reaction data workflows and provide an example of how the UDM is used to simplify the data exchange
https://acs.digitellinc.com/acs/live/21/page/411/1?eventSearchInput=&eventSearchTrack[]=83&eventSearchRoom[]=85

August 22, 2021
04:30pm - 06:30pm ET
Machine Learning & AI for Organic Chemistry Symposium  [w/ abstracts]
Organizer: Connor Coley (MIT)
This symposium will focus on the application of machine learning techniques to the understanding and prediction of chemical reactivity. Specific topics might include data-driven catalyst design, reaction product prediction, reaction yield prediction, optimization of reaction conditions, reactive force fields.
https://acs.digitellinc.com/acs/live/21/page/411/1?eventSearchInput=&eventSearchTrack[]=83&eventSearchTag[]=253&eventSearchRoom[]=70

04:55pm - 05:15pm ET
NameRxn: More than just a reaction classifier
Ingvar Lagerstedt, Roger Sayle; John Mayfield, NextMove Software Ltd, Cambridge, UK
NameRxn was developed to recognise and categorise reactions from their connection tables [from… journal or patent data. ]  The recently added reactions together with improvements to existing reaction rules and data extraction increased the classification rate from ~ 66 to ~ 73 % on reactions extracted from patents (Pistachio). …
https://acs.digitellinc.com/acs/live/21/page/411/1?eventSearchInput=&eventSearchTrack[]=83&eventSearchTag[]=253&eventSearchRoom[]=70