We present an ultrafast neural network (NN) model, QLKNN, which predicts core tokamak transport heat and particle fluxes. QLKNN is a surrogate model based on a database of 300 million flux calculations of the quasilinear gyrokinetic transport model QuaLiKiz. The database covers a wide range of realistic tokamak core parameters. Physical features such as the existence of a critical gradient for the onset of turbulent transport were integrated into the neural network training methodology. We have coupled QLKNN to the tokamak modelling framework JINTRAC and rapid control-oriented tokamak transport solver RAPTOR. The coupled frameworks are demonstrated and validated through application to three JET shots covering a representative spread of H-mode operating space, predicting turbulent transport of energy and particles in the plasma core. JINTRAC-QLKNN and RAPTOR-QLKNN are able to accurately reproduce JINTRAC-QuaLiKiz T i,e and n e profiles, but 3 to 5 orders of magnitude faster. Simulations which take hours are reduced down to only a few tens of seconds. The discrepancy in the final source-driven predicted profiles between QLKNN and QuaLiKiz is on the order 1%-15%. Also the dynamic behaviour was well captured by QLKNN, with differences of only 4%-10% compared to JINTRAC-QuaLiKiz observed at mid-radius, for a study of density buildup following the L-H transition. Deployment of neural network surrogate models in multi-physics integrated tokamak modelling is a promising route towards enabling accurate and fast tokamak scenario optimization, Uncertainty Quantification, and control applications.

No abstract

No abstract

No abstract

The bacteriophage T4 segA gene lies in a genetically unmapped region between the gene f3gt (13-glucosyltransferase) and uvsX (recombination protein) and encodes a protein of 221 amino acids. We have found that the first 100 amino acids of the SegA protein are highly similar to the N termini of four other predicted T4 proteins, also of unknown function. Together these five proteins, SegA-E (similar to endonucleases of group I introns), contain regions of similarity to the endonuclease I-Tev I, which is encoded by the mobile group I intron of the T4 td gene, and to putative endonucleases of group I introns present in the mitochondria of Neurospora crassa, Podospora anserina, and Saccharomyces douglasi. Intron-encoded endonucleases are required for the movement (homing) of the intron DNA into an intronless gene, cutting at or near the site ofintron insertion. Our in vitro assays indicate that SegA, like I-Tev I, is a Mg2+-dependent DNA endonuclease that has preferred sites for cutting. Unlike the I-Tev I gene, however, there is no evidence that segA (or the other seg genes) resides within introns. Thus, it is possible that segA encodes an endonuclease that is involved in the movement of the endonuclease-encoding DNA rather than in the homing of an intron.Group I introns have been found in the DNA of many nonmetazoan species, including the nuclear DNA of Tetrahymena, the mitochondrial genomes of fungi, the genomes of cyanobacteria and bacteriophages, and chloroplast DNA (for reviews see refs. 1-5). Despite their locations in a wide range of organisms, these introns share similar sequence and structural features, as well as a common pathway for splicing. In addition, some group I introns are mobile, transferring at a very high frequency to their respective intronless genes and creating the precise intron-exon junctions. This "homing" movement is distinct from transposition in which an element moves to unrelated insertion sites.Intron homing was first described for w, a group I intron in the mitochondrial 21S rRNA gene of Saccharomyces cerevisiae (6). Movement of w is dependent on the endonuclease I-Sce I, which is encoded within the cl intron sequence (7,8). Other group I introns that encode specific endonucleases needed for homing include an intron in the rRNA gene of Physarum polycephalum nuclear DNA (9), two introns in the bacteriophage T4 genome (within the td and sun Ygenes) (10), a second intron in the S. cerevisiae mitochondrial DNA (11,12), and introns in Chlamydomonas chloroplast DNA (13,14). The S. cerevisiae, Physarum, and Chlamydomonas endonucleases catalyze double-strand (ds) breaks very close to the insertion site (11,12,(14)(15)(16)(17) whereas the phage enzymes cut several base pairs away (18,19). The cuts are thought to initiate the insertion of the intron into the cleaved site by ds break repair (20).Although several group I introns are competent both to home and to splice, the sequence elements necessary for splicing appear to be functionally and evolutionarily independent of the DNA that e...