| 1. |
- Ermolaev, G. A., et al.
(författare)
-
Giant optical anisotropy in transition metal dichalcogenides for next-generation photonics
- 2021
-
Ingår i: Nature Communications. - : Springer Science and Business Media LLC. - 2041-1723 .- 2041-1723. ; 12:1
-
Tidskriftsartikel (refereegranskat)abstract
- Large optical anisotropy observed in a broad spectral range is of paramount importance for efficient light manipulation in countless devices. Although a giant anisotropy has been recently observed in the mid-infrared wavelength range, for visible and near-infrared spectral intervals, the problem remains acute with the highest reported birefringence values of 0.8 in BaTiS3 and h-BN crystals. This issue inspired an intensive search for giant optical anisotropy among natural and artificial materials. Here, we demonstrate that layered transition metal dichalcogenides (TMDCs) provide an answer to this quest owing to their fundamental differences between intralayer strong covalent bonding and weak interlayer van der Waals interaction. To do this, we made correlative far- and near-field characterizations validated by first-principle calculations that reveal a huge birefringence of 1.5 in the infrared and 3 in the visible light for MoS2. Our findings demonstrate that this remarkable anisotropy allows for tackling the diffraction limit enabling an avenue for on-chip next-generation photonics. Optical anisotropy in a broad spectral range is pivotal to efficient light manipulation. Here, the authors measure a birefringence of 1.5 in the infrared range and 3 in the visible light for MoS2.
|
|
| 2. |
- Laine, Elodie, et al.
(författare)
-
Protein sequence-to-structure learning : Is this the end(-to-end revolution)?
- 2021
-
Ingår i: Proteins. - : Wiley. - 0887-3585 .- 1097-0134. ; 89:12, s. 1770-1786
-
Forskningsöversikt (refereegranskat)abstract
- The potential of deep learning has been recognized in the protein structure prediction community for some time, and became indisputable after CASP13. In CASP14, deep learning has boosted the field to unanticipated levels reaching near-experimental accuracy. This success comes from advances transferred from other machine learning areas, as well as methods specifically designed to deal with protein sequences and structures, and their abstractions. Novel emerging approaches include (i) geometric learning, that is, learning on representations such as graphs, three-dimensional (3D) Voronoi tessellations, and point clouds; (ii) pretrained protein language models leveraging attention; (iii) equivariant architectures preserving the symmetry of 3D space; (iv) use of large meta-genome databases; (v) combinations of protein representations; and (vi) finally truly end-to-end architectures, that is, differentiable models starting from a sequence and returning a 3D structure. Here, we provide an overview and our opinion of the novel deep learning approaches developed in the last 2 years and widely used in CASP14.
|
|
| 3. |
- Menéndez Hurtado, David, 1990-
(författare)
-
Structured Learning for Structural Bioinformatics : Applications of Deep Learning to Protein Structure Prediction
- 2019
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Proteins are the basic molecular machines of the cell, performing a broad range of tasks, from structural support to catalysisof chemical reactions. Their function is determined by their 3D structure, which in turn is dictated by the order of their components, the amino acids.This thesis is dedicated to applications of machine learning to the problems of contact prediction, ab-initio, and model quality assessment. In particular, my research has been focused on developing methods that are both effective, and easy to use.In the first paper, we improved the already state-of-the-art model quality assessment (MQA) program ProQ3 replacing the underlying machine learning algorithm from svm to Deep Learning, baptised ProQ3D. The correlation between predicted and true scores was improved from 0.85 to 0.90, using the same training data and features.The second paper joined several programs into a single pipeline for ab-initio structure prediction: contact prediction,folding, and model selection. We attempted to predict the structures of all 6379 PFAM families with unknown structure, ofwhich 558 we believe to be accurate. Of these, 415 had not been reported before.The third paper uses advances in machine learning to build a contact predictor, PconsC4, that is fast and easy to deployin large-scale studies, since it requires a single Multiple Sequence Alignment (MSA), and no external dependencies. The predictions are state-of-the-art, yielding a 12% improvement in precision over PconsC3, and 244 times faster.With ProQ4, in the fourth paper, we introduce a novel way of training deep networks for MQA in a way that minimises the bias of the training data, and emphasises model ranking, and demonstrate its viability with a minimal description ofthe protein. The ranking correlation was improved with respect to ProQ3D from 0.82 to 0.90.Lastly, in the fifth paper, weshow the results of ProQ3D and ProQ4 in a completely blind test: CASP13.
|
|
| 4. |
- Moretti, Rocco, et al.
(författare)
-
Community-wide evaluation of methods for predicting the effect of mutations on protein-protein interactions
- 2013
-
Ingår i: Proteins. - : Wiley. - 0887-3585 .- 1097-0134. ; 81:11, s. 1980-1987
-
Tidskriftsartikel (refereegranskat)abstract
- Community-wide blind prediction experiments such as CAPRI and CASP provide an objective measure of the current state of predictive methodology. Here we describe a community-wide assessment of methods to predict the effects of mutations on protein-protein interactions. Twenty-two groups predicted the effects of comprehensive saturation mutagenesis for two designed influenza hemagglutinin binders and the results were compared with experimental yeast display enrichment data obtained using deep sequencing. The most successful methods explicitly considered the effects of mutation on monomer stability in addition to binding affinity, carried out explicit side-chain sampling and backbone relaxation, evaluated packing, electrostatic, and solvation effects, and correctly identified around a third of the beneficial mutations. Much room for improvement remains for even the best techniques, and large-scale fitness landscapes should continue to provide an excellent test bed for continued evaluation of both existing and new prediction methodologies. Proteins 2013; 81:1980-1987.
|
|
| 5. |
- Wahlgren, J., et al.
(författare)
-
Gene segment reassortment between American and Asian lineages of avian influenza virus from waterfowl in the Beringia area
- 2008
-
Ingår i: Vector Borne and Zoonotic Diseases. - : Mary Ann Liebert Inc. - 1530-3667 .- 1557-7759. ; 8:6, s. 783-90
-
Tidskriftsartikel (refereegranskat)abstract
- Since prehistoric times, the Bering Strait area (Beringia) has served as an avenue of dispersal between the Old and the New Worlds. On a field expedition to this area, we collected fecal samples from dabbling ducks, geese, shorebirds, and gulls on the Chukchi Peninsula, Siberia, and Pt. Barrow, Alaska, and characterized the subtypes of avian influenza virus present in them. Four of 202 samples (2%) from Alaska were positive for influenza A virus RNA in two independent polymerase chain reaction (PCR)-based screening assays, while all shorebird samples from the Chukchi Peninsula were negative. Subtypes H3N8 and H6N1 were recorded once, while subtype H8N4 was found in two samples. Full-length sequences were obtained from the three unique isolates, and phylogenetic analysis with representative sequences for the Eurasian and North American lineages of influenza A virus showed that one HA gene clustered with the Eurasian rather than the North American lineage. However, the closest relative to this sequence was a North American isolate from Delaware described in 2002, indicating that a H6 spillover from Asia has established itself in North America.
|
|
