Particle picking is a time-consuming step in single-particle analysis and often requires significant interventions from users, which has become a bottleneck for future automated electron cryo-microscopy (cryo-EM). Here we report a deep learning framework, called DeepPicker, to address this problem and fill the current gaps toward a fully automated cryo-EM pipeline. DeepPicker employs a novel cross-molecule training strategy to capture common features of particles from previously-analyzed micrographs, and thus does not require any human intervention during particle picking. Tests on the recently-published cryo-EM data of three complexes have demonstrated that our deep learning based scheme can successfully accomplish the human-level particle picking process and identify a sufficient number of particles that are comparable to those manually by human experts. These results indicate that DeepPicker can provide a practically useful tool to significantly reduce the time and manual effort spent in single-particle analysis and thus greatly facilitate high-resolution cryo-EM structure determination.
Broadband near-infrared (NIR) light sources based on phosphorconverted light-emitting diodes (pc-LEDs) are desirable for biochemical analysis and medical diagnosis applications; however, the development of target NIR phosphor is still a challenge. Herein, broadband NIR phosphors, Cr 3+ -activated CaSc 1−x Al 1+x SiO 6 (λ em = 950 nm), are designed and optimized by chemical substitution toward enhanced quantum efficiency and thermal stability. Structural and spectral analyses along with density functional theory calculations reveal that Sc 3+ /Al 3+ substitution contributes to enhancing the structural rigidity and the local symmetry of the [Sc/AlO 6 ] octahedron so that the nonradiative relaxation of Cr 3+ emission centers is suppressed significantly. The as-fabricated phosphor-in-glassbased NIR LED light source demonstrates great potential in the detection of alcohol concentration. This study provides a local structure design principle for exploring NIR phosphors with enhanced thermal stability and will also stimulate further studies on material discovery and quantitative analysis of NIR spectroscopy.
Summary
Near-infrared (NIR) phosphor-converted light-emitting diode (pc-LED) has great potential in non-invasive detection, while the discovery of tunable broadband NIR phosphor still remains a challenge. Here, we report that Cr
3+
-activated LiIn
2
SbO
6
exhibits a broad emission band ranging from 780 to 1400 nm with a full width at half maximum (FWHM) of 225 nm upon 492 nm excitation. The emission peaks are tuned from 970 to 1020 nm together with considerable broadening of FWHM (∼285 nm) via Li/Na substitution. Depending on Yb
3+
co-doping, a stronger NIR fluorescence peak of Yb
3+
appears with improved thermal resistance, which is ascribed to efficient energy transfer from Cr
3+
to Yb
3+
. An NIR pc-LED package has been finally designed and demonstrated a remarkable ability to penetrate pork tissues (∼2 cm) so that the insertion depth of a needle can be observed, indicating that the phosphor can be applied in non-destructive monitoring.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.