For the normative version of our publication list see Christoph Steinbeck‘s ORCID profile.
2000
Krause, Stefan; Willighagen, Egon; Steinbeck, Christoph
JChemPaint - Using the Collaborative Forces of the Internet to Develop a Free Editor for 2D Chemical Structures Journal Article
In: Molecules, vol. 5, no. 1, pp. 93–98, 2000.
@article{krause2000jchempaint,
title = {JChemPaint - Using the Collaborative Forces of the Internet to Develop a Free Editor for 2D Chemical Structures},
author = {Krause, Stefan and Willighagen, Egon and Steinbeck, Christoph},
url = {http://www.mdpi.org/molecules/papers/50100093.pdf},
year = {2000},
date = {2000-01-01},
journal = {Molecules},
volume = {5},
number = {1},
pages = {93--98},
publisher = {Molecular Diversity Preservation International},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
1999
Ho, W C; Steinbeck, C; Richert, C
Solution structure of the aminoacyl-capped oligodeoxyribonucleotide duplex (W-TGCGCAC)(2) Journal Article
In: Biochemistry, vol. 38, no. 39, pp. 12597–12606, 1999.
@article{ho1999solution,
title = {Solution structure of the aminoacyl-capped oligodeoxyribonucleotide duplex (W-TGCGCAC)(2)},
author = {Ho, W C and Steinbeck, C and Richert, C},
url = {http://dx.doi.org/10.1021/bi991169w},
doi = {10.1021/bi991169w},
year = {1999},
date = {1999-01-01},
journal = {Biochemistry},
volume = {38},
number = {39},
pages = {12597--12606},
publisher = {ACS Publications},
abstract = {Reported here is the solution structure of the aminoacyl-DNA duplex (W-TGCGCAC)(2). This duplex forms a continuously pi-stacked helix consisting of both nucleobases and amino acid side chains. According to NMR and UV analyses, the duplex melts in a cooperative transition and with 1.3-1.8% greater hyperchromicity than the control duplex (TGCGCAC)(2). A van't Hoff analysis of UV melting points at different concentrations shows that the two tryptophan residues contribute 4.8 kcal/mol to the Delta H degrees of complex formation at 10 mM salt concentration and less than 1 kcal/mol at 150 mM I salt. The entropic cost for duplex association in the presence of the amino acid residues is 13 cal/molK greater than that for the control at 10 mM salt concentration, and 3 cal/molK lower than that of the control at 0.15 ionic strength. The conformation of W-TGCGCAC in duplex form, determined via restrained torsion angle molecular dynamics, shows an undisturbed B-form DNA duplex with dangling 3'-termini. The tryptophanyl residue at the 5'-terminus packs tightly against T2 and the proximal part of adenine, without engaging in hydrogen bonding. While not providing strong enthalpic net stabilization of the duplex, the tryptophan "cap" on the duplex does seem to reduce the fraying at the termini, indicating a subtle balance of entropic and enthalpic factors contributing to the molecular dynamics. The structure also shows that, at least in the present sequence context, stacking on the terminal base pair is more favorable than intercalation, probably because the enthalpic cost associated with breaking up the stacking between DNA base pairs cannot be paid for by favorable pi-stacking interactions with the indole ring of tryptophan. These results are of importance for understanding stacking interactions in protein-DNA complexes, particularly those in enzyme-substrate complexes involving exposed nucleobases. [References: 69]},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Reported here is the solution structure of the aminoacyl-DNA duplex (W-TGCGCAC)(2). This duplex forms a continuously pi-stacked helix consisting of both nucleobases and amino acid side chains. According to NMR and UV analyses, the duplex melts in a cooperative transition and with 1.3-1.8% greater hyperchromicity than the control duplex (TGCGCAC)(2). A van't Hoff analysis of UV melting points at different concentrations shows that the two tryptophan residues contribute 4.8 kcal/mol to the Delta H degrees of complex formation at 10 mM salt concentration and less than 1 kcal/mol at 150 mM I salt. The entropic cost for duplex association in the presence of the amino acid residues is 13 cal/molK greater than that for the control at 10 mM salt concentration, and 3 cal/molK lower than that of the control at 0.15 ionic strength. The conformation of W-TGCGCAC in duplex form, determined via restrained torsion angle molecular dynamics, shows an undisturbed B-form DNA duplex with dangling 3'-termini. The tryptophanyl residue at the 5'-terminus packs tightly against T2 and the proximal part of adenine, without engaging in hydrogen bonding. While not providing strong enthalpic net stabilization of the duplex, the tryptophan "cap" on the duplex does seem to reduce the fraying at the termini, indicating a subtle balance of entropic and enthalpic factors contributing to the molecular dynamics. The structure also shows that, at least in the present sequence context, stacking on the terminal base pair is more favorable than intercalation, probably because the enthalpic cost associated with breaking up the stacking between DNA base pairs cannot be paid for by favorable pi-stacking interactions with the indole ring of tryptophan. These results are of importance for understanding stacking interactions in protein-DNA complexes, particularly those in enzyme-substrate complexes involving exposed nucleobases. [References: 69]
1998
Zhang, G L; Rucker, G; Breitmeier, E; Mayer, R; Steinbeck, C
Alkaloids from Thalictrum przewalskii. Journal Article
In: Planta Medica, vol. 64, no. 2, pp. 165–171, 1998.
@article{Zhang:1998ci,
title = {Alkaloids from Thalictrum przewalskii.},
author = {Zhang, G L and Rucker, G and Breitmeier, E and Mayer, R and Steinbeck, C},
url = {http://www.thieme-connect.de/DOI/DOI?10.1055/s-2006-957396},
doi = {10.1055/s-2006-957396},
year = {1998},
date = {1998-01-01},
journal = {Planta Medica},
volume = {64},
number = {2},
pages = {165--171},
publisher = {textcopyright Georg Thieme Verlag Stuttgart textperiodcentered New York},
abstract = {Nine new alkaloids przewaline, przewalskine, przewalskinine, przewalstine, przewalstinine, przewalstidine, przewalstidinine, przewalidine chloride, and 8-hydroxypseudocoptisine chloride were isolated from Thalictrum przewalskii Maxim., together with the known alkaloids berberinium chloride, magnoflorine chloride, N-methylpalaudinium chloride, thalphenine chloride, and N-methylnantenium chloride. Their structures were determined by spectroscopy including EI-MS, FAB-MS, (1)H-NMR, (13)C-NMR, C,H-COSY, C,H-COLOC, HMQC and HMBC techniques.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Nine new alkaloids przewaline, przewalskine, przewalskinine, przewalstine, przewalstinine, przewalstidine, przewalstidinine, przewalidine chloride, and 8-hydroxypseudocoptisine chloride were isolated from Thalictrum przewalskii Maxim., together with the known alkaloids berberinium chloride, magnoflorine chloride, N-methylpalaudinium chloride, thalphenine chloride, and N-methylnantenium chloride. Their structures were determined by spectroscopy including EI-MS, FAB-MS, (1)H-NMR, (13)C-NMR, C,H-COSY, C,H-COLOC, HMQC and HMBC techniques.
Steinbeck, Christoph; Richert, Clemens
The Role of Ionic Backbones in RNA Structure:~ An Unusually Stable Non-WatsontextminusCrick Duplex of a Nonionic Analog in an Apolar Medium Journal Article
In: Journal of the American Chemical Society, vol. 120, no. 45, pp. 11576–11580, 1998.
@article{Steinbeck:1998fj,
title = {The Role of Ionic Backbones in RNA Structure:~ An Unusually Stable Non-WatsontextminusCrick Duplex of a Nonionic Analog in an Apolar Medium},
author = {Christoph Steinbeck and Clemens Richert},
url = {https://pubs.acs.org/doi/10.1021/ja9817951},
doi = {10.1021/ja9817951},
year = {1998},
date = {1998-01-01},
journal = {Journal of the American Chemical Society},
volume = {120},
number = {45},
pages = {11576--11580},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
1997
Steinbeck, C; Spitzer, V; Starosta, M; von Poser, G
Identification of two chromenes from Calea serrata by semiautomatic structure elucidation Journal Article
In: Journal of Natural Products, vol. 60, no. 6, pp. 627–628, 1997.
@article{steinbeck1997identification,
title = {Identification of two chromenes from Calea serrata by semiautomatic structure elucidation},
author = {Steinbeck, C and Spitzer, V and Starosta, M and von Poser, G},
url = {http://dx.doi.org/10.1021/np960742z},
doi = {10.1021/np960742z},
year = {1997},
date = {1997-01-01},
journal = {Journal of Natural Products},
volume = {60},
number = {6},
pages = {627--628},
publisher = {ACS Publications},
abstract = {The isolation and semiautomatic structure identification of the two chromenes, eupatoriochromene (1) and preconene II (2), from the aerial parts of Calea serrata is described. The structure elucidation was performed on the basis of 1D and 2D NMR methods using the recently published computer program LUCY. Neither compound has been isolated from Calea serrata previously. [References: 5]},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The isolation and semiautomatic structure identification of the two chromenes, eupatoriochromene (1) and preconene II (2), from the aerial parts of Calea serrata is described. The structure elucidation was performed on the basis of 1D and 2D NMR methods using the recently published computer program LUCY. Neither compound has been isolated from Calea serrata previously. [References: 5]
Berlin, K; Jain, R K; Tetzlaff, C; Steinbeck, C; Richert, C
Spectrometrically monitored selection experiments - quantitative laser desorption mass spectrometry of small chemical libraries Journal Article
In: Chemistry & Biology, vol. 4, no. 1, pp. 63–77, 1997.
@article{berlin1997spectrometrically,
title = {Spectrometrically monitored selection experiments - quantitative laser desorption mass spectrometry of small chemical libraries},
author = {Berlin, K and Jain, R K and Tetzlaff, C and Steinbeck, C and Richert, C},
url = {http://dx.doi.org/10.1016/S1074-5521(97)90237-4},
doi = {10.1016/S1074-5521(97)90237-4},
year = {1997},
date = {1997-01-01},
journal = {Chemistry & Biology},
volume = {4},
number = {1},
pages = {63--77},
publisher = {Elsevier},
abstract = {Background: Selection experiments involving chemical libraries are routinely used in the pharmaceutical industry for finding and optimizing lead compounds, In principle, almost any process involving a binding event or a reaction could be probed systematically with chemical libraries prepared by combinatorial synthesis, Traditionally, however, the vast majority of library members cannot be monitored during the selection, making a systematic correlation of structure and activity difficult, To interpret selection experiments on the level of all library components, monitoring technologies are required that give a unique and quantitative spectroscopic signal for every compound in a mixture. Results: Quantitative matrix-assisted laser desorption mass spectrometry of libraries of porphyrins and peptide-DNA hybrids consisting of 2-35 compounds is described. Porphyrin libraries were subjected to in vitro selections for liposome incorporation and binding to a protein pocket, it was shown that meso-hydroxyphenyl substituted porphyrins, known high activity photosensitizers of tumors, are preferentially incorporated in liposome membranes. A mixture of peptide-DNA hybrids was assayed for the nuclease stability of its components. Conclusions: Small libraries of non-isobaric compounds can be exhaustively or near-exhaustively monitored by mass spectrometry. Monitored selection experiments can yield detailed structure-activity maps in a single experiment, speeding up drug discovery and the probing of biochemically relevant recognition events. It is proposed that monitored assays for target binding, membrane partitioning, and biostability could be run in parallel, to select drug candidates combining several favorable properties in 'multidimensional' selection experiments. [References: 42]},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Background: Selection experiments involving chemical libraries are routinely used in the pharmaceutical industry for finding and optimizing lead compounds, In principle, almost any process involving a binding event or a reaction could be probed systematically with chemical libraries prepared by combinatorial synthesis, Traditionally, however, the vast majority of library members cannot be monitored during the selection, making a systematic correlation of structure and activity difficult, To interpret selection experiments on the level of all library components, monitoring technologies are required that give a unique and quantitative spectroscopic signal for every compound in a mixture. Results: Quantitative matrix-assisted laser desorption mass spectrometry of libraries of porphyrins and peptide-DNA hybrids consisting of 2-35 compounds is described. Porphyrin libraries were subjected to in vitro selections for liposome incorporation and binding to a protein pocket, it was shown that meso-hydroxyphenyl substituted porphyrins, known high activity photosensitizers of tumors, are preferentially incorporated in liposome membranes. A mixture of peptide-DNA hybrids was assayed for the nuclease stability of its components. Conclusions: Small libraries of non-isobaric compounds can be exhaustively or near-exhaustively monitored by mass spectrometry. Monitored selection experiments can yield detailed structure-activity maps in a single experiment, speeding up drug discovery and the probing of biochemically relevant recognition events. It is proposed that monitored assays for target binding, membrane partitioning, and biostability could be run in parallel, to select drug candidates combining several favorable properties in 'multidimensional' selection experiments. [References: 42]
Steinbeck, C; Berlin, K; Richert, C
Masp - a Program Predicting Mass Spectra of Combinatorial Libraries Journal Article
In: Journal of Chemical Information & Computer Sciences, vol. 37, no. 3, pp. 449–457, 1997.
@article{steinbeck1997masp,
title = {Masp - a Program Predicting Mass Spectra of Combinatorial Libraries},
author = {Steinbeck, C and Berlin, K and Richert, C},
url = {http://dx.doi.org/10.1021/ci960160n},
doi = {10.1021/ci960160n},
year = {1997},
date = {1997-01-01},
journal = {Journal of Chemical Information & Computer Sciences},
volume = {37},
number = {3},
pages = {449--457},
publisher = {ACS Publications},
abstract = {MASP, a program predicting fragmentation-free mass spectra of libraries prepared by combinatorial synthesis, is presented. MASP combines user-defined building blocks with a nonvariable core molecule and calculates isotopically resolved mass spectra. Peak overlap and the abundance of major and minor isotope peaks can be examined interactively and read from a color display of the spectrum. Further, MASP exhaustively screens a Meta-Library Space to identify libraries with minimum peak overlap and maximum diversity. Diversity can be defined by the user, e.g., as size, hydrophobicity, or hydrogen bonding capability. The usefulness of such a screen for the design of libraries is demonstrated for meso-substituted tetraphenylporphyrins and peptides with a meta-library space of up to 3080 libraries and 625 compounds per library. The program operates on a PC platform under MS Windows 95. It may be useful for drug discovery and optimization studies that employ methods of combinatorial synthesis and mass spectrometry-guided in vitro selection. [References: 16]},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
MASP, a program predicting fragmentation-free mass spectra of libraries prepared by combinatorial synthesis, is presented. MASP combines user-defined building blocks with a nonvariable core molecule and calculates isotopically resolved mass spectra. Peak overlap and the abundance of major and minor isotope peaks can be examined interactively and read from a color display of the spectrum. Further, MASP exhaustively screens a Meta-Library Space to identify libraries with minimum peak overlap and maximum diversity. Diversity can be defined by the user, e.g., as size, hydrophobicity, or hydrogen bonding capability. The usefulness of such a screen for the design of libraries is demonstrated for meso-substituted tetraphenylporphyrins and peptides with a meta-library space of up to 3080 libraries and 625 compounds per library. The program operates on a PC platform under MS Windows 95. It may be useful for drug discovery and optimization studies that employ methods of combinatorial synthesis and mass spectrometry-guided in vitro selection. [References: 16]
1996
Steinbeck, C
LUCY-ein Programm zur Konstitutionsermittlung aus NMR-Korrelationsexperimenten Mein łdots Journal Article
In: Angewandte Chemie, 1996.
@article{Steinbeck:1996vs,
title = {LUCY-ein Programm zur Konstitutionsermittlung aus NMR-Korrelationsexperimenten Mein łdots},
author = {Steinbeck, C},
url = {http://www3.interscience.wiley.com/journal/112392792/abstract},
year = {1996},
date = {1996-01-01},
journal = {Angewandte Chemie},
abstract = {9525-9533. Korrelationsexperimenten** wurde speziell zur Verwendung von HMBC-Daten entworfen.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
9525-9533. Korrelationsexperimenten** wurde speziell zur Verwendung von HMBC-Daten entworfen.
Berlin, K; Steinbeck, C; Breitmaier, E
Synthesis of carba-porphyrinoids from tripyrranes and unsaturated dialdehydes Journal Article
In: Synthesis-Stuttgart, no. 3, pp. 336 ff., 1996.
@article{Berlin1996,
title = {Synthesis of carba-porphyrinoids from tripyrranes and unsaturated dialdehydes},
author = {Berlin, K and Steinbeck, C and Breitmaier, E},
year = {1996},
date = {1996-01-01},
journal = {Synthesis-Stuttgart},
number = {3},
pages = {336 ff.},
abstract = {Hitherto unknown triaza[18]annulenes 3-5 and 7-10 are synthesized by cyclocondensation of tripyrranes 2 with azulene-1,3-dicarboxaldehyde (1) or cycloheptatriene-1,6-dicarboxaldehyde (6). Their spectroscopical properties with respect to 18 pi aromaticity are discussed. [References: 26]},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Hitherto unknown triaza[18]annulenes 3-5 and 7-10 are synthesized by cyclocondensation of tripyrranes 2 with azulene-1,3-dicarboxaldehyde (1) or cycloheptatriene-1,6-dicarboxaldehyde (6). Their spectroscopical properties with respect to 18 pi aromaticity are discussed. [References: 26]
Steinbeck, C
Lucy - A Program For Structure Elucidation From NMR Correlation Experiments Journal Article
In: Angewandte Chemie International Edition in English, vol. 35, no. 17, pp. 1984–1986, 1996.
@article{steinbeck1996lucy,
title = {Lucy - A Program For Structure Elucidation From NMR Correlation Experiments},
author = {Steinbeck, C},
url = {http://www.google.de/search?client=safari&rls=10_7_4&q=Lucy+A+Program+For+Structure+Elucidation+From+NMR+Correlation+Experiments&ie=UTF-8&oe=UTF-8&redir_esc=&ei=Mvm5UIOoLMrm4QT1ooEw},
year = {1996},
date = {1996-01-01},
journal = {Angewandte Chemie International Edition in English},
volume = {35},
number = {17},
pages = {1984--1986},
publisher = {Wiley Online Library},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
1995
Steinbeck, C; Schneider, C; Rotscheidt, K; Breitmaier, E
A 4-methyl-7-hydroxyphthalide glycoside and other constituents from Quillaja saponaria molina. Journal Article
In: Phytochemistry, vol. 40, no. 4, pp. 1313–1315, 1995.
@article{Steinbeck:1995ve,
title = {A 4-methyl-7-hydroxyphthalide glycoside and other constituents from Quillaja saponaria molina.},
author = {Steinbeck, C and Schneider, C and Rotscheidt, K and Breitmaier, E},
url = {http://eutils.ncbi.nlm.nih.gov/entrez/eutils/elink.fcgi?dbfrom=pubmed&id=7492375&retmode=ref&cmd=prlinks},
year = {1995},
date = {1995-11-01},
journal = {Phytochemistry},
volume = {40},
number = {4},
pages = {1313--1315},
abstract = {A so far unknown 4-methyl-7-hydroxyphthalideglycoside has been isolated from the methanol extract of the bark of Quillaja saponaria molina. Its structure has been established from NMR experiments as 7-O(-)[beta-glucopyranosyl-(1-->6)-beta-arabinopyranosyl]-7-hydrox y-4-methy l -1[3H]-isobenzofuranone. Two known compounds, 3,4,5-trimethoxyphenyl-beta-D-glucopyranoside and lyoniresinol-3 alpha-O-beta-D-glycopyranoside were also identified.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
A so far unknown 4-methyl-7-hydroxyphthalideglycoside has been isolated from the methanol extract of the bark of Quillaja saponaria molina. Its structure has been established from NMR experiments as 7-O(-)[beta-glucopyranosyl-(1-->6)-beta-arabinopyranosyl]-7-hydrox y-4-methy l -1[3H]-isobenzofuranone. Two known compounds, 3,4,5-trimethoxyphenyl-beta-D-glucopyranoside and lyoniresinol-3 alpha-O-beta-D-glycopyranoside were also identified.
Zhang, G L; Rucker, G; Breitmaier, E; Nieger, M; Mayer, R; Steinbeck, C
Alkaloids from Dactylicapnos torulosa Journal Article
In: Phytochemistry, vol. 40, no. 1, pp. 299–305, 1995.
@article{zhang1995alkaloids,
title = {Alkaloids from Dactylicapnos torulosa},
author = {Zhang, G L and Rucker, G and Breitmaier, E and Nieger, M and Mayer, R and Steinbeck, C},
url = {http://dx.doi.org/10.1016/0031-9422(95)00192-A},
doi = {10.1016/0031-9422(95)00192-A},
year = {1995},
date = {1995-01-01},
journal = {Phytochemistry},
volume = {40},
number = {1},
pages = {299--305},
publisher = {Elsevier},
abstract = {Phytochemical investigation of Dactylicapnos torulosa yielded two known compounds,(-)-cis-N-methyl-stylopiumchloride and hydrastinine chloride; and five new alkaloids, dactyline, 8-hydroxydihydrosanguinarine and dactylidine, as well as dactylicapnosine and dactylicapnosinine with novel C-N-O-C moieties. All structures were elucidated by spectroscopical methods. The structure of dactylicapnosine was also determined by single crystal X-ray diffraction analysis. [References: 18]},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Phytochemical investigation of Dactylicapnos torulosa yielded two known compounds,(-)-cis-N-methyl-stylopiumchloride and hydrastinine chloride; and five new alkaloids, dactyline, 8-hydroxydihydrosanguinarine and dactylidine, as well as dactylicapnosine and dactylicapnosinine with novel C-N-O-C moieties. All structures were elucidated by spectroscopical methods. The structure of dactylicapnosine was also determined by single crystal X-ray diffraction analysis. [References: 18]
0000
Rajan, Kohulan; Zielesny, Achim; Steinbeck, Christoph
DECIMER 1.0: deep learning for chemical image recognition using transformers Journal Article
In: J Cheminform, vol. 13, no. 1, pp. 61, 0000, ISSN: 1758-2946.
@article{pmid34404468,
title = {DECIMER 1.0: deep learning for chemical image recognition using transformers},
author = {Kohulan Rajan and Achim Zielesny and Christoph Steinbeck},
doi = {10.1186/s13321-021-00538-8},
issn = {1758-2946},
journal = {J Cheminform},
volume = {13},
number = {1},
pages = {61},
abstract = {The amount of data available on chemical structures and their properties has increased steadily over the past decades. In particular, articles published before the mid-1990 are available only in printed or scanned form. The extraction and storage of data from those articles in a publicly accessible database are desirable, but doing this manually is a slow and error-prone process. In order to extract chemical structure depictions and convert them into a computer-readable format, Optical Chemical Structure Recognition (OCSR) tools were developed where the best performing OCSR tools are mostly rule-based. The DECIMER (Deep lEarning for Chemical ImagE Recognition) project was launched to address the OCSR problem with the latest computational intelligence methods to provide an automated open-source software solution. Various current deep learning approaches were explored to seek a best-fitting solution to the problem. In a preliminary communication, we outlined the prospect of being able to predict SMILES encodings of chemical structure depictions with about 90% accuracy using a dataset of 50-100 million molecules. In this article, the new DECIMER model is presented, a transformer-based network, which can predict SMILES with above 96% accuracy from depictions of chemical structures without stereochemical information and above 89% accuracy for depictions with stereochemical information.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The amount of data available on chemical structures and their properties has increased steadily over the past decades. In particular, articles published before the mid-1990 are available only in printed or scanned form. The extraction and storage of data from those articles in a publicly accessible database are desirable, but doing this manually is a slow and error-prone process. In order to extract chemical structure depictions and convert them into a computer-readable format, Optical Chemical Structure Recognition (OCSR) tools were developed where the best performing OCSR tools are mostly rule-based. The DECIMER (Deep lEarning for Chemical ImagE Recognition) project was launched to address the OCSR problem with the latest computational intelligence methods to provide an automated open-source software solution. Various current deep learning approaches were explored to seek a best-fitting solution to the problem. In a preliminary communication, we outlined the prospect of being able to predict SMILES encodings of chemical structure depictions with about 90% accuracy using a dataset of 50-100 million molecules. In this article, the new DECIMER model is presented, a transformer-based network, which can predict SMILES with above 96% accuracy from depictions of chemical structures without stereochemical information and above 89% accuracy for depictions with stereochemical information.