Predicting the outcomes of organic reactions via machine learning: are current descriptors sufficient?

Predicting the outcomes of organic reactions via machine learning: are current descriptors sufficient?. Skoraczyński, G, Dittwald, P, Miasojedow, B, Szymkuć, S, Gajewska, E P, Grzybowski, B A, & Gambin, A Scientific Reports, 7(1):3582, 2017.

Paper doi abstract bibtex

As machine learning/artificial intelligence algorithms are defeating chess masters and, most recently, GO champions, there is interest – and hope – that they will prove equally useful in assisting chemists in predicting outcomes of organic reactions. This paper demonstrates, however, that the applicability of machine learning to the problems of chemical reactivity over diverse types of chemistries remains limited – in particular, with the currently available chemical descriptors, fundamental mathematical theorems impose upper bounds on the accuracy with which raction yields and times can be predicted. Improving the performance of machine-learning methods calls for the development of fundamentally new chemical descriptors.

@article{skoraczynski_predicting_2017,
	title = {Predicting the outcomes of organic reactions via machine learning: are current descriptors sufficient?},
	volume = {7},
	issn = {2045-2322},
	url = {https://doi.org/10.1038/s41598-017-02303-0},
	doi = {10.1038/s41598-017-02303-0},
	abstract = {As machine learning/artificial intelligence algorithms are defeating chess masters and, most recently, GO champions, there is interest – and hope – that they will prove equally useful in assisting chemists in predicting outcomes of organic reactions. This paper demonstrates, however, that the applicability of machine learning to the problems of chemical reactivity over diverse types of chemistries remains limited – in particular, with the currently available chemical descriptors, fundamental mathematical theorems impose upper bounds on the accuracy with which raction yields and times can be predicted. Improving the performance of machine-learning methods calls for the development of fundamentally new chemical descriptors.},
	number = {1},
	journal = {Scientific Reports},
	author = {Skoraczyński, G and Dittwald, P and Miasojedow, B and Szymkuć, S and Gajewska, E P and Grzybowski, B A and Gambin, A},
	year = {2017},
	pages = {3582},
}

Downloads: 0

{"_id":"pSbrL9poEbGgA8A6y","bibbaseid":"skoraczyski-dittwald-miasojedow-szymku-gajewska-grzybowski-gambin-predictingtheoutcomesoforganicreactionsviamachinelearningarecurrentdescriptorssufficient-2017","author_short":["Skoraczyński, G","Dittwald, P","Miasojedow, B","Szymkuć, S","Gajewska, E P","Grzybowski, B A","Gambin, A"],"bibdata":{"bibtype":"article","type":"article","title":"Predicting the outcomes of organic reactions via machine learning: are current descriptors sufficient?","volume":"7","issn":"2045-2322","url":"https://doi.org/10.1038/s41598-017-02303-0","doi":"10.1038/s41598-017-02303-0","abstract":"As machine learning/artificial intelligence algorithms are defeating chess masters and, most recently, GO champions, there is interest – and hope – that they will prove equally useful in assisting chemists in predicting outcomes of organic reactions. This paper demonstrates, however, that the applicability of machine learning to the problems of chemical reactivity over diverse types of chemistries remains limited – in particular, with the currently available chemical descriptors, fundamental mathematical theorems impose upper bounds on the accuracy with which raction yields and times can be predicted. Improving the performance of machine-learning methods calls for the development of fundamentally new chemical descriptors.","number":"1","journal":"Scientific Reports","author":[{"propositions":[],"lastnames":["Skoraczyński"],"firstnames":["G"],"suffixes":[]},{"propositions":[],"lastnames":["Dittwald"],"firstnames":["P"],"suffixes":[]},{"propositions":[],"lastnames":["Miasojedow"],"firstnames":["B"],"suffixes":[]},{"propositions":[],"lastnames":["Szymkuć"],"firstnames":["S"],"suffixes":[]},{"propositions":[],"lastnames":["Gajewska"],"firstnames":["E","P"],"suffixes":[]},{"propositions":[],"lastnames":["Grzybowski"],"firstnames":["B","A"],"suffixes":[]},{"propositions":[],"lastnames":["Gambin"],"firstnames":["A"],"suffixes":[]}],"year":"2017","pages":"3582","bibtex":"@article{skoraczynski_predicting_2017,\n\ttitle = {Predicting the outcomes of organic reactions via machine learning: are current descriptors sufficient?},\n\tvolume = {7},\n\tissn = {2045-2322},\n\turl = {https://doi.org/10.1038/s41598-017-02303-0},\n\tdoi = {10.1038/s41598-017-02303-0},\n\tabstract = {As machine learning/artificial intelligence algorithms are defeating chess masters and, most recently, GO champions, there is interest – and hope – that they will prove equally useful in assisting chemists in predicting outcomes of organic reactions. This paper demonstrates, however, that the applicability of machine learning to the problems of chemical reactivity over diverse types of chemistries remains limited – in particular, with the currently available chemical descriptors, fundamental mathematical theorems impose upper bounds on the accuracy with which raction yields and times can be predicted. Improving the performance of machine-learning methods calls for the development of fundamentally new chemical descriptors.},\n\tnumber = {1},\n\tjournal = {Scientific Reports},\n\tauthor = {Skoraczyński, G and Dittwald, P and Miasojedow, B and Szymkuć, S and Gajewska, E P and Grzybowski, B A and Gambin, A},\n\tyear = {2017},\n\tpages = {3582},\n}\n\n","author_short":["Skoraczyński, G","Dittwald, P","Miasojedow, B","Szymkuć, S","Gajewska, E P","Grzybowski, B A","Gambin, A"],"key":"skoraczynski_predicting_2017","id":"skoraczynski_predicting_2017","bibbaseid":"skoraczyski-dittwald-miasojedow-szymku-gajewska-grzybowski-gambin-predictingtheoutcomesoforganicreactionsviamachinelearningarecurrentdescriptorssufficient-2017","role":"author","urls":{"Paper":"https://doi.org/10.1038/s41598-017-02303-0"},"metadata":{"authorlinks":{}},"html":""},"bibtype":"article","biburl":"https://bibbase.org/zotero/kjelljorner","dataSources":["64RFiGdCa5JLg5kLH"],"keywords":[],"search_terms":["predicting","outcomes","organic","reactions","via","machine","learning","current","descriptors","sufficient","skoraczyński","dittwald","miasojedow","szymkuć","gajewska","grzybowski","gambin"],"title":"Predicting the outcomes of organic reactions via machine learning: are current descriptors sufficient?","year":2017}