1 of 16

CHEMISTRY IN TIMES OF ARTIFICIAL INTELLIGENCE

Asia Cicmil

Andjelija Djoric

2 of 16

INTRODUCTION

LEARNING IN CHEMISTRY

DATABASES

THE BARRIERS TO ADOPTING AI IN CHEMISTRY

ANALYTICAL CHEMISTRY

TABLE OF CONTENTS

01

02

04

05

03

3 of 16

INTRODUCTION

01

Does anyone know what AI stands for?

4 of 16

Artificial Intelligence(AI) is the single biggest technology revolution the world has ever seen

It has entered many domains of society. Chemists have to a large extent gained their knowledge by doing experiments and thus gather data.

TABLE OF CONTENTS

5 of 16

LEARNING IN CHEMISTRY

02

6 of 16

TABLE OF CONTENTS

7 of 16

DATABASES

03

8 of 16

In the beginning, various forms of computer-readable chemical structure representations were explored as a basis for processing chemical structures and reactions and for building databases

​

With the rapid development of computer technology computer storage space became more easily and cheaper available

​

This allowed the coding of chemical structures in a manner that opened many desirable possibilities for structure processing and manipulations

TABLE OF CONTENTS

DATABASES

9 of 16

THE BARRIERS TO ADOPTING AI IN CHEMISTRY

04

10 of 16

Three key barriers to adopting AI in chemistry:

  1. Data quality Optimal predictions are dependent upon high quality datasets that provide both positive and negative examples for training
  2. Technology While improvements are being made in computing power there are still perceived limitations from a user perspective
  3. Talent shortages Increasing collaboration between chemistry and other scientific disciplines may help accelerate the integration of AI

TABLE OF CONTENTS

11 of 16

ANALYTICAL CHEMISTRY

05

12 of 16

Analytical chemistry is the science of obtaining, processing, and communicating information about the composition and structure of matter

​

The characterization of chemical objects is the domain of analytical chemistry. The objects can be compounds, samples from archeology, food samples, explosives, medical plants, urine samples, etc. These objects are investigated by a variety of methods such as chromatography, spectroscopy, etc. generating a host of data

ANALYTICAL CHEMISTRY

13 of 16

Picture shows the results of a study for finding out from which of nine areas a sample of an Italian olive oil came from. Each olive oil was characterized by its content of eight different fatty acids. 250 samples of the available 572 samples were used to train a self-organizing neural network (SONN). From the additional 322 samples, 312 could correctly be predicted by the SONN.

​

OLIVE OIL STUDY

14 of 16

Artificial intelligence translates chemistry to predict reaction outcomes

15 of 16

RESOURCES

f

​

CREDITS: This presentation template was created by Slidesgo, including icons by Flaticon, and infographics & images by Freepik

​

​

16 of 16

Thank you for your attention!