ترغب بنشر مسار تعليمي؟ اضغط هنا

Molecule Identification with Rotational Spectroscopy and Probabilistic Deep Learning

97   0   0.0 ( 0 )
 نشر من قبل Kin Long Kelvin Lee
 تاريخ النشر 2020
  مجال البحث فيزياء
والبحث باللغة English




اسأل ChatGPT حول البحث

A proof-of-concept framework for identifying molecules of unknown elemental composition and structure using experimental rotational data and probabilistic deep learning is presented. Using a minimal set of input data determined experimentally, we describe four neural network architectures that yield information to assist in the identification of an unknown molecule. The first architecture translates spectroscopic parameters into Coulomb matrix eigenspectra, as a method of recovering chemical and structural information encoded in the rotational spectrum. The eigenspectrum is subsequently used by three deep learning networks to constrain the range of stoichiometries, generate SMILES strings, and predict the most likely functional groups present in the molecule. In each model, we utilize dropout layers as an approximation to Bayesian sampling, which subsequently generates probabilistic predictions from otherwise deterministic models. These models are trained on a modestly sized theoretical dataset comprising ${sim}$83,000 unique organic molecules (between 18 and 180 amu) optimized at the $omega$B97X-D/6-31+G(d) level of theory where the theoretical uncertainty of the spectroscopic constants are well understood and used to further augment training. Since chemical and structural properties depend highly on molecular composition, we divided the dataset into four groups corresponding to pure hydrocarbons, oxygen-bearing, nitrogen-bearing, and both oxygen- and nitrogen-bearing species, training each type of network with one of these categories thus creating experts within each domain of molecules. We demonstrate how these models can then be used for practical inference on four molecules, and discuss both the strengths and shortcomings of our approach, and the future directions these architectures can take.



قيم البحث

اقرأ أيضاً

We demonstrate experimentally a method of all-optical selective rotational control in gas mixtures. Using an optical centrifuge - an intense laser pulse whose linear polarization rotates at an accelerated rate, we simultaneously excite two different molecular species to two different rotational frequencies of choice. The new level of control is achieved by shaping the centrifuge spectrum according to the rotational spectra of the centrifuged molecules. The shaped optical centrifuge releases one molecular species earlier than the other, therefore separating their target rotational frequencies and corresponding rotational states. The technique will expand the utility of rotational control in the studies of the effects of molecular rotation on collisions and chemical reactions.
Glycans play a central role as mediators in most biological processes, but their structures are complicated by isomerism. Epimers and anomers, regioisomers, and branched sequences contribute to a structural variability that dwarfs those of nucleic ac ids and proteins, challenging even the most sophisticated analytical tools, such as NMR and mass spectrometry. Here, we introduce an electron tunneling technique that is label-free and can identify carbohydrates at the single-molecule level, offering significant benefits over existing technology. It is capable of analyzing sub-picomole quantities of sample, counting the number of individual molecules in each subset in a population of coexisting isomers, and is quantitative over more than four orders of magnitude of concentration. It resolves epimers not well separated by ion-mobility and can be implemented on a silicon chip. It also provides a readout mechanism for direct single-molecule sequencing of linear oligosaccharides.
Hypericin can be found in nature in Hypericum perforatum (St. Johns Wort) and has become subject of intense biochemical research. Studies report of antidepressive, antineoplastic, antitumor and antiviral activity of hypericin. Among the variety of po tential applications hypericin can be used as photosensitizer in photodynamic therapy (PDT), where it is brought into cancer cells and produces singlet oxygen upon irradiation with a suitable light source. Therefore, the photophysical properties of hypericin are crucial for a successful application in a medical treatment. Here, we present the first single molecule optical spectroscopy study of hypericin. Its photostability is large enough to obtain single molecule fluorescence, surface enhanced Raman spectra (SERS), fluorescence lifetime, antibunching and blinking dynamics. Embedding hypericin in a PVA matrix changes the blinking dynamics, reduces the fluorescence lifetime and increases the photostability. Single molecule SERS spectra show both the neutral and deprotonated form of hypericin and exhibit sudden spectral changes, which can be associated with a reorientation of the single molecule with respect to the surface.
Single molecule spectroscopy aims at unveiling often hidden but potentially very important contributions of single entities to a systems ensemble response. Albeit contributing tremendously to our ever growing understanding of molecular processes the fundamental question of temporal evolution, or change, has thus far been inaccessible, resulting in a static picture of a dynamic world. Here, we finally resolve this dilemma by performing the first ultrafast time-resolved transient spectroscopy on a single molecule. By tracing the femtosecond evolution of excited electronic state spectra of single molecules over hundreds of nanometres of bandwidth at room temperature we reveal their non-linear ultrafast response in an effective 3-pulse scheme with fluorescence detection. A first excitation pulse is followed by a phase-locked de-excitation pulse-pair, providing spectral encoding with 25 fs temporal resolution. This experimental realisation of true single molecule transient spectroscopy demonstrates that two-dimensional electronic spectroscopy of single molecules is experimentally in reach.
Knowledge Transfer (KT) techniques tackle the problem of transferring the knowledge from a large and complex neural network into a smaller and faster one. However, existing KT methods are tailored towards classification tasks and they cannot be used efficiently for other representation learning tasks. In this paper a novel knowledge transfer technique, that is capable of training a student model that maintains the same amount of mutual information between the learned representation and a set of (possible unknown) labels as the teacher model, is proposed. Apart from outperforming existing KT techniques, the proposed method allows for overcoming several limitations of existing methods providing new insight into KT as well as novel KT applications, ranging from knowledge transfer from handcrafted feature extractors to {cross-modal} KT from the textual modality into the representation extracted from the visual modality of the data.
التعليقات
جاري جلب التعليقات جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
mircosoft-partner

هل ترغب بارسال اشعارات عن اخر التحديثات في شمرا-اكاديميا