_:b9325908 _:b9325912 . _:b9325908 _:b9325913 . . _:b9325908 _:b9325914 . _:b448200425 . _:b9325908 _:b9325915 . _:b448200424 . _:b448200427 . _:b448200340 . _:b448200884 . _:b9325923 "The combinatorial libraries were synthesized as previously described [>>24<<,41]. Peptides in each library are 13-mers with Alanine residues in positions 1, 2, 12 and 13." . _:b448200426 . _:b448200653 . _:b448200437 . _:b448200615 . . _:b448200376 . _:b448200436 . _:b448200872 . _:b448200439 . . . _:b448200438 . . _:b448200433 . . _:b448200432 . _:b448200422 . _:b448200435 . _:b448200434 . _:b9325952 . _:b448200445 . _:b448200631 . . _:b448200444 . _:b448200447 . _:b448200446 . _:b9325926 . _:b448200441 . _:b448200440 . _:b448200489 . _:b448200443 . _:b448200772 . _:b448200442 . . . _:b9325953 . _:b448200342 . . . _:b448200765 . _:b448200687 . _:b448200718 . _:b448200733 . _:b448200822 . _:b9325930 "Here we adopted the similarity measure described by El-Manzalawy et al. [>>32<<]. Two peptides were defined as similar if they satisfied one of the following conditions:" . _:b448200514 . _:b448200672 . . . _:b9325975 "We plan to systematically examine how to best construct consensus predictions for MHC binding in the future, building on work done by us and others in the past [30,>>39<<,40]." . _:b448200442 . . _:b448200861 . . _:b9325927 . . . _:b448200327 . . _:b448200648 . . . . _:b9325957 . . _:b448200671 . _:b9325919 "Clearly, reducing sequence similarity had a significant impact on the observed classifier performance, which is consistent with previous findings [>>32<<]. At the same time, the order of performance of the different prediction methods was unchanged when using the reduced dataset, with NN-align performing the best, SMM-align second, ARB third, and PROPRED and the combinatorial libraries" . _:b448200853 . _:b448200435 . _:b448200443 . _:b9325933 "For binding assays, the peptides were classified into binders (experimental IC50 < 1000 nM) and nonbinders (experimental IC50\u22651000 nM) as described previously [>>30<<]. For a given prediction method and a given cutoff for the predicted scores, the rate of true positive and false positive predictions can be calculated. An ROC curve is generated by varying the cutoff from the highest to the lowest" . . . _:b448200494 . . _:b9325961 "discussion and conclusions" . . _:b448200763 . _:b448200663 . . _:b9325941 "Both alpha and beta chains can impact the distinct peptide binding specificity of an HLA class II molecule [>>6<<]. HLA class II peptide ligands that are recognized by T cells and trigger an immune response are referred to as immune epitopes [7]. Identifying such epitopes can help detect and modulate immune responses in infectious diseases, allergy," . . _:b448200633 . _:b448200418 . _:b9325917 "The second new method we added to the IEDB analysis resource was NN-align [>>25<<]. This method differs from previous approaches in that NN-align is neural network based and can hence take into account higher order sequence correlations." . . . . _:b448200449 . . _:b448200428 . _:b448200508 . . . _:b448200485 . _:b448200626 . _:b448200445 . . _:b448200857 . _:b448200461 . _:b9325962 . . _:b448200565 . _:b448200355 . _:b448200606 . _:b9325912 "This uniformly large number of more than 500 affinity measurements for each included allelic variant was previously found to be required to consistently generate reliable predictions [>>30<<]. To the best of our knowledge, this is the first publicly available dataset of HLA-DP and HLA-DQ binding affinities of significant size." . _:b448200429 . _:b448200534 . . . _:b448200474 . . . _:b448200352 . _:b9325929 . . . . _:b448200742 . . _:b448200666 . . _:b448200868 . _:b448200369 . _:b448200783 . _:b9325962 . _:b9325913 "Among those methods, ARB, SMM-align and PROPRED (based on the matrices constructed by Sturniolo et al. [>>16<<] on which also the TEPITOPE predictions are based) were the top performing ones and were incorporated into the MHC class II binding prediction component of the IEDB analysis resource [21]." . _:b448200846 . _:b448200424 . _:b448200802 . _:b9325941 . . . _:b448200462 . _:b448200816 . . _:b448200704 . _:b448200800 . _:b448200426 . . . . . . _:b448200591 . _:b9325962 . . _:b448200757 . . . . _:b448200360 . _:b9325974 "We plan to systematically examine how to best construct consensus predictions for MHC binding in the future, building on work done by us and others in the past [>>30<<,39,40]." . . . . . . . . . . . _:b448200341 . . _:b448200542 . . _:b448200647 . . _:b448200551 . _:b448200782 . . _:b9325944 . _:b9325959 "This neural network based approach combines the peptide sequence representation used in the NetMHC algorithm [26,27] that was highly successful in predicting the binding specificity of HLA class I molecules [28,>>29<<] with the representation of peptide flanking residues and peptide length used in NetMHCIIpan method [19]." . _:b448200804 . _:b448200548 . . . . _:b448200794 . _:b448200343 . . _:b448200880 . _:b448200715 . . _:b448200574 . _:b448200405 . _:b448200466 . _:b448200337 . . . _:b448200712 . . . . . _:b448200869 . _:b448200330 . . . . . _:b448200779 . _:b448200673 . . . . . . . . _:b9325970 "This repeats the dominating performance of the related NetMHC prediction methods in a number of recent MHC class I prediction benchmarks [>>28<<,29,38]." . . _:b448200450 . _:b448200440 . . . . . . . _:b9325954 "This approach is also time and cost effective, as the same panel of peptide libraries can be scanned for all HLA molecules of interest, and has been applied successfully for multiple applications [>>22<<-24], The second method we newly implemented was NN-align [25]." . _:b448200478 . . _:b448200828 . . . _:b9325961 . _:b9325914 . . . . _:b448200578 . _:b9325922 . _:b448200769 . _:b448200392 . _:b448200749 . _:b448200759 . . _:b9325935 . . _:b448200509 . . . _:b448200751 . . . . _:b9325908 . . . _:b9325920 "This was demonstrated in [>>32<<] in which a classifier displayed better performance than others when evaluated on a dataset that contained similar sequences, but which completely failed when evaluated on a dataset with no homology between peptides." . . . _:b448200821 . _:b448200641 . . _:b448200586 . . . _:b448200870 . . _:b448200587 . . . _:b9325945 "Several databases have been established to document the results of such experiments including Antijen [9], MHCBN [>>10<<], MHCPEP [11], FIMM [12], SYFPEITHI [13] and the Immune Epitope Database (IEDB) [14,15]." . . . _:b448200719 . . . . _:b9325938 "The molecules are heterodimers consisting of an alpha chain and a beta chain encoded in one of three loci: HLA DR, DP and DQ [2,>>3<<]. The DR locus can encode two beta chains DRB1 and DRB3-5 which are in linkage disequilibrium [4]. The genes encoding class II molecules are highly polymorphic, as evidenced by the IMGT/HLA database [5] which lists 1,190 known sequences" . _:b448200446 . . _:b448200676 . . _:b9325951 . _:b448200473 . _:b448200388 . _:b448200722 . . . . _:b448200419 . _:b448200731 . _:b448200554 . . _:b448200616 . . . _:b448200608 . . . _:b448200453 . . _:b448200837 . _:b9325911 "Secondly, for several molecules, relatively few data points existed, in spite of the fact that we and others [30,>>31<<] have shown that several hundred data points are desirable to derive accurate predictive algorithms." . _:b448200827 . _:b448200867 . _:b448200482 . _:b448200581 . _:b448200572 . . _:b448200717 . _:b448200707 . . . . . . _:b9325928 "Several previous studies have proposed measurements to determine peptide similarity [>>32<<,46-49]. Here we adopted the similarity measure described by El-Manzalawy et al. [32]." . . _:b448200890 . _:b448200762 . . _:b448200448 . _:b448200412 . . _:b448200881 . . . . . . _:b448200391 . . _:b448200834 . _:b448200305 "21"^^ . _:b9325959 . _:b448200307 "18"^^ . _:b448200304 "24"^^ . _:b448200634 . _:b9325925 . _:b448200306 "19"^^ . _:b448200790 . . _:b9325937 . . . _:b448200308 "18"^^ . . _:b448200309 "17"^^ . _:b448200303 "51"^^ . . _:b448200310 "16"^^ . _:b9325917 . _:b448200311 "16"^^ . _:b448200469 . _:b448200312 "16"^^ . _:b448200313 "15"^^ . _:b448200314 "15"^^ . . . _:b448200315 "15"^^ . _:b448200331 . _:b448200316 "14"^^ . _:b448200703 . . _:b9325971 . _:b448200317 "14"^^ . _:b448200470 . . _:b448200318 "14"^^ . _:b448200319 "14"^^ . _:b448200327 "12"^^ . _:b448200326 "12"^^ . _:b448200325 "12"^^ . _:b448200444 . _:b448200324 "12"^^ . _:b448200439 . _:b448200323 "12"^^ . _:b448200322 "12"^^ . _:b448200321 "12"^^ . . _:b448200320 "13"^^ . _:b448200335 "10"^^ . _:b448200334 "10"^^ . _:b9325934 "The plotting of ROC curve and calculation of AUC were carried out with the ROCR [>>53<<] package for R [54]." . _:b448200525 . . _:b448200333 "11"^^ . _:b448200332 "11"^^ . _:b448200848 . _:b448200331 "11"^^ . _:b448200330 "11"^^ . _:b9325957 "This neural network based approach combines the peptide sequence representation used in the NetMHC algorithm [26,>>27<<] that was highly successful in predicting the binding specificity of HLA class I molecules [28,29] with the representation of peptide flanking residues and peptide length used in NetMHCIIpan method [19]." . _:b448200329 "12"^^ . _:b448200856 . _:b448200328 "12"^^ . . . _:b448200342 "8"^^ . . _:b448200343 "8"^^ . _:b448200340 "9"^^ . _:b9325909 "In a previous report, we described the release of 10,017 MHC class II binding affinities experimentally measured by our group [>>30<<]. The data included measured binding affinities for a total of 17 different mouse and human allelic variants." . _:b448200341 "8"^^ . . . _:b448200338 "9"^^ . _:b448200339 "9"^^ . . _:b448200336 "10"^^ . _:b448200337 "9"^^ . _:b448200350 "7"^^ . . . _:b448200351 "7"^^ . . _:b448200681 . _:b448200492 . _:b448200348 "7"^^ . . . _:b448200349 "7"^^ . _:b448200346 "7"^^ . . . _:b448200347 "7"^^ . _:b9325970 . "10.1186%2F1471-2105-11-568" . _:b448200320 . _:b448200344 "8"^^ . _:b448200801 . _:b448200345 "7"^^ . . . _:b448200358 "6"^^ . . . _:b448200359 "6"^^ . _:b448200356 "7"^^ . _:b448200709 "2"^^ . _:b448200357 "7"^^ . . _:b448200736 . _:b448200354 "7"^^ . _:b448200708 "2"^^ . _:b448200709 . _:b448200577 . _:b448200355 "7"^^ . _:b448200738 . _:b448200711 "2"^^ . _:b448200539 . _:b448200352 "7"^^ . . _:b448200710 "2"^^ . . _:b448200538 . _:b448200882 . _:b448200353 "7"^^ . _:b448200705 "2"^^ . _:b448200366 "6"^^ . _:b448200704 "2"^^ . _:b448200402 . _:b9325955 . _:b448200367 "6"^^ . _:b448200707 "2"^^ . _:b448200732 . _:b9325958 . _:b448200364 "6"^^ . _:b448200706 "2"^^ . _:b448200365 "6"^^ . _:b448200717 "2"^^ . _:b448200362 "6"^^ . _:b448200716 "2"^^ . . _:b448200363 "6"^^ . _:b448200719 "2"^^ . . _:b448200360 "6"^^ . . _:b448200718 "2"^^ . _:b448200361 "6"^^ . _:b448200729 . _:b448200713 "2"^^ . . _:b9325973 . _:b448200712 "2"^^ . _:b9325964 . . _:b448200373 "5"^^ . _:b9325949 "Several databases have been established to document the results of such experiments including Antijen [9], MHCBN [10], MHCPEP [11], FIMM [12], SYFPEITHI [13] and the Immune Epitope Database (IEDB) [>>14<<,15]. IEDB currently documents 12,577 peptides tested for binding to one of more of 158 MHC class II allelic variants of which 114 are human (HLA). It is possible to develop binding prediction methods for HLA molecules for which no" . _:b448200715 "2"^^ . _:b9325963 . _:b448200372 "5"^^ . _:b448200714 "2"^^ . _:b448200375 "5"^^ . _:b448200725 "2"^^ . _:b448200374 "5"^^ . . . _:b448200724 "2"^^ . _:b448200335 . _:b448200590 . _:b448200727 "2"^^ . _:b448200368 "6"^^ . _:b448200726 "2"^^ . _:b448200371 "5"^^ . _:b448200369 "6"^^ . . _:b448200721 "2"^^ . _:b448200370 "5"^^ . _:b448200720 "2"^^ . _:b448200381 "5"^^ . _:b448200752 . _:b448200723 "2"^^ . _:b448200506 . _:b448200380 "5"^^ . _:b9325958 "This neural network based approach combines the peptide sequence representation used in the NetMHC algorithm [26,27] that was highly successful in predicting the binding specificity of HLA class I molecules [>>28<<,29] with the representation of peptide flanking residues and peptide length used in NetMHCIIpan method [19]." . _:b448200722 "2"^^ . _:b448200383 "5"^^ . . _:b448200733 "2"^^ . _:b448200382 "5"^^ . _:b448200732 "2"^^ . _:b448200377 "5"^^ . _:b448200735 "2"^^ . _:b448200376 "5"^^ . _:b448200582 . _:b448200734 "2"^^ . _:b448200379 "5"^^ . _:b448200873 . _:b9325946 "Several databases have been established to document the results of such experiments including Antijen [9], MHCBN [10], MHCPEP [>>11<<], FIMM [12], SYFPEITHI [13] and the Immune Epitope Database (IEDB) [14,15]." . _:b448200729 "2"^^ . _:b448200378 "5"^^ . _:b448200728 "2"^^ . _:b448200389 "5"^^ . . _:b448200731 "2"^^ . . _:b448200388 "5"^^ . . _:b448200730 "2"^^ . _:b448200391 "5"^^ . _:b448200741 "2"^^ . _:b448200390 "5"^^ . _:b448200740 "2"^^ . _:b448200818 . _:b448200579 . _:b448200385 "5"^^ . _:b448200743 "2"^^ . _:b448200609 . _:b448200384 "5"^^ . _:b448200742 "2"^^ . _:b9325974 . _:b448200387 "5"^^ . _:b448200737 "2"^^ . _:b448200386 "5"^^ . _:b448200736 "2"^^ . _:b448200397 "4"^^ . . . _:b448200739 "2"^^ . . . _:b448200396 "5"^^ . _:b448200738 "2"^^ . _:b448200399 "4"^^ . _:b9325911 . _:b448200749 "2"^^ . _:b448200398 "4"^^ . _:b448200680 . _:b448200748 "2"^^ . _:b448200393 "5"^^ . _:b448200751 "2"^^ . _:b448200392 "5"^^ . _:b448200750 "2"^^ . _:b448200500 . . _:b448200395 "5"^^ . _:b448200501 . _:b448200502 . _:b448200745 "2"^^ . _:b9325933 . _:b448200503 . _:b448200496 . _:b448200394 "5"^^ . _:b448200497 . _:b448200498 . _:b448200744 "2"^^ . _:b448200499 . _:b448200405 "4"^^ . _:b448200508 . . _:b448200509 . _:b448200510 . _:b448200747 "2"^^ . _:b448200468 . _:b448200407 . _:b448200511 . _:b448200404 "4"^^ . _:b448200504 . _:b448200505 . _:b448200506 . _:b448200746 "2"^^ . _:b448200507 . _:b448200407 "4"^^ . _:b9325918 "One such approach is to remove similar peptides from the entire dataset [>>32<<]. We call this a 'random selection' strategy as the order in which peptides are removed is not defined." . _:b448200484 . _:b448200485 . . _:b448200486 . _:b448200757 "2"^^ . _:b448200487 . _:b448200406 "4"^^ . _:b448200480 . _:b448200481 . _:b448200482 . _:b448200756 "2"^^ . _:b448200483 . _:b448200401 "4"^^ . _:b448200492 . _:b448200493 . _:b448200494 . _:b448200759 "2"^^ . _:b448200495 . _:b448200400 "4"^^ . _:b448200488 . _:b448200489 . _:b448200490 . _:b9325924 . _:b448200758 "2"^^ . _:b9325943 . _:b448200491 . _:b448200403 "4"^^ . _:b448200468 . . _:b448200469 . _:b448200470 . _:b448200753 "2"^^ . _:b448200471 . _:b448200402 "4"^^ . _:b448200464 . _:b448200465 . _:b448200466 . _:b448200752 "2"^^ . _:b448200467 . _:b448200413 "4"^^ . _:b448200476 . _:b448200477 . _:b448200478 . _:b448200755 "2"^^ . _:b448200479 . . _:b448200412 "4"^^ . _:b448200472 . _:b448200473 . . _:b448200474 . _:b448200754 "2"^^ . _:b448200475 . _:b448200415 "4"^^ . _:b448200452 . _:b448200453 . _:b448200454 . _:b448200456 . _:b448200455 . _:b448200765 "2"^^ . _:b448200414 "4"^^ . _:b448200448 . _:b448200449 . . _:b448200450 . _:b448200764 "2"^^ . _:b448200420 . _:b448200451 . _:b448200409 "4"^^ . _:b448200460 . _:b448200322 . _:b448200461 . _:b448200462 . . _:b448200767 "2"^^ . . _:b448200463 . _:b448200408 "4"^^ . _:b448200456 . _:b448200457 . _:b448200458 . _:b448200766 "2"^^ . . _:b448200459 . _:b448200411 "4"^^ . _:b448200564 . _:b448200565 . _:b448200566 . _:b448200761 "2"^^ . _:b448200567 . _:b448200773 . _:b448200560 . _:b448200410 "4"^^ . _:b448200561 . _:b448200562 . _:b448200760 "2"^^ . _:b448200563 . _:b448200421 "4"^^ . _:b448200572 . _:b448200573 . _:b448200574 . _:b448200763 "2"^^ . . _:b448200575 . _:b448200420 "4"^^ . _:b448200568 . _:b448200569 . . _:b448200570 . _:b448200625 . _:b448200762 "2"^^ . _:b448200571 . _:b448200423 "4"^^ . _:b448200548 . _:b448200549 . _:b448200550 . _:b448200773 "2"^^ . _:b448200551 . _:b448200422 "4"^^ . _:b448200544 . _:b448200545 . _:b448200546 . . _:b448200546 . _:b448200772 "2"^^ . _:b448200318 . _:b448200547 . _:b448200417 "4"^^ . . _:b448200556 . _:b448200557 . _:b448200558 . . _:b448200775 "2"^^ . _:b448200559 . _:b448200416 "4"^^ . _:b448200777 . _:b448200404 . . . _:b448200552 . _:b448200553 . _:b448200554 . _:b448200774 "2"^^ . _:b448200555 . _:b448200419 "4"^^ . _:b448200532 . _:b448200533 . _:b448200534 . _:b448200769 "2"^^ . _:b448200535 . _:b448200418 "4"^^ . _:b448200528 . _:b448200529 . _:b448200530 . _:b448200768 "2"^^ . _:b448200531 . _:b448200429 "4"^^ . _:b9325929 . _:b448200540 . _:b448200541 . _:b448200542 . _:b448200771 "2"^^ . _:b448200543 . _:b448200428 "4"^^ . _:b448200536 . _:b448200537 . _:b448200538 . _:b448200567 . _:b448200539 . _:b448200770 "2"^^ . _:b448200431 "4"^^ . _:b448200516 . . _:b448200517 . _:b448200518 . _:b448200781 "2"^^ . . _:b448200519 . _:b448200315 . _:b448200512 . _:b448200430 "4"^^ . _:b448200513 . _:b448200514 . _:b448200780 "2"^^ . _:b448200745 . _:b448200515 . _:b448200425 "4"^^ . . _:b448200524 . _:b448200525 . _:b448200526 . _:b448200783 "2"^^ . _:b448200527 . _:b448200424 "4"^^ . _:b448200520 . _:b448200562 . _:b448200521 . _:b448200522 . _:b448200782 "2"^^ . _:b448200523 . _:b448200427 "4"^^ . . _:b448200628 . _:b448200629 . _:b448200630 . _:b448200777 "2"^^ . _:b448200631 . _:b448200426 "4"^^ . _:b9325953 "It is our goal to include a variety of binding prediction algorithms in the IEDB Analysis Resource (IEDB-AR) [>>21<<], identify the best performing ones, and ideally combine multiple algorithms into a superior consensus prediction." . _:b448200624 . _:b448200625 . _:b448200626 . _:b448200437 "3"^^ . _:b448200627 . _:b448200776 "2"^^ . _:b448200636 . _:b448200637 . _:b448200535 . _:b448200638 . _:b448200436 "3"^^ . _:b448200639 . _:b448200779 "2"^^ . . _:b448200632 . _:b448200633 . _:b448200634 . _:b448200439 "3"^^ . _:b448200635 . _:b448200778 "2"^^ . _:b448200612 . _:b448200613 . _:b448200614 . _:b448200438 "3"^^ . _:b448200615 . _:b448200789 "2"^^ . _:b448200608 . _:b448200695 . _:b448200609 . _:b448200610 . _:b448200788 "2"^^ . _:b448200611 . _:b448200433 "4"^^ . _:b448200620 . _:b448200621 . _:b448200622 . _:b448200791 "2"^^ . _:b448200623 . _:b448200432 "4"^^ . _:b448200616 . _:b448200617 . _:b448200618 . _:b448200435 "3"^^ . _:b448200619 . _:b448200790 "2"^^ . . _:b448200596 . _:b448200500 . _:b448200597 . _:b448200598 . _:b448200434 "3"^^ . _:b448200599 . _:b448200785 "2"^^ . _:b448200592 . . _:b448200593 . _:b448200594 . _:b448200445 "3"^^ . _:b448200595 . _:b448200784 "2"^^ . _:b448200604 . . _:b448200605 . _:b448200606 . . _:b448200444 "3"^^ . _:b448200607 . _:b448200787 "2"^^ . _:b448200600 . _:b448200601 . _:b448200602 . _:b448200447 "3"^^ . . _:b448200603 . _:b448200786 "2"^^ . _:b448200580 . _:b448200581 . _:b448200582 . _:b448200446 "3"^^ . _:b448200583 . _:b448200797 "2"^^ . _:b448200576 . . _:b448200577 . _:b448200578 . _:b448200441 "3"^^ . _:b448200579 . _:b448200796 "2"^^ . _:b448200588 . _:b448200589 . _:b448200590 . _:b448200440 "3"^^ . _:b448200591 . _:b448200799 "2"^^ . _:b448200584 . _:b448200585 . _:b448200586 . _:b448200443 "3"^^ . _:b448200587 . _:b448200798 "2"^^ . . _:b448200692 . _:b448200693 . _:b448200694 . _:b448200442 "3"^^ . _:b448200695 . _:b448200793 "2"^^ . _:b448200543 . _:b448200688 . _:b448200505 . _:b448200689 . _:b448200690 . _:b448200453 "3"^^ . _:b9325916 . _:b448200691 . _:b448200792 "2"^^ . _:b448200700 . _:b448200701 . . _:b448200702 . _:b448200452 "3"^^ . _:b9325915 . _:b448200703 . _:b448200795 "2"^^ . _:b448200696 . _:b448200697 . _:b448200698 . _:b448200455 "3"^^ . _:b448200726 . _:b448200699 . _:b448200794 "2"^^ . _:b448200785 . _:b448200676 . _:b448200677 . _:b448200400 . _:b448200678 . _:b448200454 "3"^^ . _:b448200679 . _:b448200805 "2"^^ . _:b448200672 . . _:b448200673 . _:b448200674 . _:b448200427 . . _:b9325912 . _:b448200675 . _:b448200804 "2"^^ . _:b448200449 "3"^^ . _:b448200684 . _:b448200685 . . _:b448200686 . _:b448200448 "3"^^ . _:b448200463 . _:b448200687 . _:b448200807 "2"^^ . _:b448200680 . _:b448200681 . _:b448200888 . _:b448200682 . _:b448200451 "3"^^ . _:b9325910 . _:b448200683 . _:b448200806 "2"^^ . _:b448200660 . _:b448200661 . _:b448200662 . _:b448200450 "3"^^ . _:b9325909 . _:b448200663 . _:b448200801 "2"^^ . _:b448200656 . _:b448200657 . . _:b448200658 . _:b448200461 "3"^^ . _:b448200659 . _:b448200800 "2"^^ . _:b448200668 . _:b448200334 . _:b448200669 . _:b448200670 . _:b448200460 "3"^^ . _:b448200671 . _:b448200803 "2"^^ . _:b448200664 . _:b448200665 . _:b448200408 . _:b448200666 . _:b448200463 "3"^^ . _:b448200667 . _:b448200802 "2"^^ . _:b448200644 . . . _:b448200645 . _:b448200646 . _:b448200462 "3"^^ . _:b448200647 . _:b448200813 "2"^^ . . _:b448200358 . _:b448200640 . _:b448200641 . _:b448200595 . _:b448200642 . _:b448200457 "3"^^ . _:b448200643 . _:b448200812 "2"^^ . _:b448200652 . _:b448200347 . . _:b448200653 . _:b448200654 . _:b448200456 "3"^^ . _:b448200655 . _:b448200815 "2"^^ . _:b448200648 . _:b448200649 . _:b448200650 . _:b448200459 "3"^^ . _:b448200651 . _:b448200814 "2"^^ . . _:b448200840 . _:b448200454 . _:b448200610 . _:b448200458 "3"^^ . _:b448200809 "2"^^ . _:b448200348 . _:b448200469 "3"^^ . _:b448200808 "2"^^ . _:b448200627 . _:b448200468 "3"^^ . _:b448200811 "2"^^ . _:b448200755 . _:b448200471 "3"^^ . _:b448200810 "2"^^ . _:b448200470 "3"^^ . _:b448200821 "2"^^ . _:b448200465 "3"^^ . _:b448200820 "2"^^ . _:b448200464 "3"^^ . _:b448200823 "2"^^ . . _:b9325940 . . . _:b448200467 "3"^^ . _:b448200822 "2"^^ . . _:b448200466 "3"^^ . _:b448200817 "2"^^ . _:b448200477 "3"^^ . _:b448200816 "2"^^ . _:b448200476 "3"^^ . _:b448200819 "2"^^ . _:b448200865 . _:b448200740 . _:b448200818 "2"^^ . _:b448200479 "3"^^ . _:b448200854 . _:b448200670 . _:b448200478 "3"^^ . _:b9325954 . _:b448200829 "2"^^ . _:b448200473 "3"^^ . _:b448200828 "2"^^ . _:b448200576 . _:b448200472 "3"^^ . _:b448200831 "2"^^ . _:b448200475 "3"^^ . _:b448200830 "2"^^ . _:b448200308 . _:b448200309 . _:b448200310 . _:b448200474 "3"^^ . _:b448200311 . _:b448200825 "2"^^ . _:b448200843 . _:b448200304 . _:b448200305 . _:b448200306 . _:b448200485 "3"^^ . . _:b448200307 . _:b448200824 "2"^^ . _:b448200316 . _:b448200317 . _:b448200318 . _:b448200484 "3"^^ . . _:b448200319 . _:b448200827 "2"^^ . . _:b448200312 . _:b448200313 . _:b448200314 . _:b448200487 "3"^^ . _:b448200315 . _:b448200826 "2"^^ . . _:b448200486 "3"^^ . _:b448200837 "2"^^ . _:b448200375 . _:b448200481 "3"^^ . _:b448200613 . _:b448200836 "2"^^ . _:b448200710 . . . . _:b448200303 . _:b448200839 "2"^^ . _:b448200480 "3"^^ . _:b448200483 "3"^^ . _:b448200838 "2"^^ . _:b448200482 "3"^^ . _:b448200833 "2"^^ . . _:b448200493 "3"^^ . _:b9325967 . _:b448200832 "2"^^ . _:b448200492 "3"^^ . _:b448200629 . _:b448200835 "2"^^ . . _:b448200495 "3"^^ . _:b448200834 "2"^^ . _:b448200494 "3"^^ . _:b448200845 "2"^^ . . _:b448200489 "3"^^ . _:b448200844 "2"^^ . _:b448200488 "3"^^ . _:b448200847 "2"^^ . _:b448200491 "3"^^ . _:b448200874 . _:b448200846 "2"^^ . _:b448200372 . _:b448200373 . _:b448200374 . _:b448200490 "3"^^ . _:b448200885 . _:b448200375 . _:b448200841 "2"^^ . _:b448200368 . _:b448200369 . _:b448200370 . _:b448200501 "3"^^ . _:b448200371 . _:b448200840 "2"^^ . _:b448200380 . _:b448200381 . _:b448200382 . _:b448200500 "3"^^ . _:b448200383 . _:b448200843 "2"^^ . _:b448200376 . _:b448200377 . _:b448200378 . _:b448200503 "3"^^ . _:b448200379 . _:b448200842 "2"^^ . _:b448200356 . _:b448200357 . _:b448200358 . . _:b448200359 . _:b448200853 "2"^^ . _:b448200502 "3"^^ . _:b448200352 . . _:b448200353 . _:b448200354 . _:b448200497 "3"^^ . _:b448200355 . _:b448200852 "2"^^ . _:b448200364 . _:b448200365 . _:b448200366 . _:b448200496 "3"^^ . _:b448200367 . _:b448200855 "2"^^ . . _:b448200360 . _:b448200713 . _:b448200361 . _:b448200362 . _:b448200499 "3"^^ . . _:b448200363 . _:b448200854 "2"^^ . _:b448200396 . _:b448200340 . . _:b448200341 . _:b448200342 . _:b448200498 "3"^^ . _:b448200343 . _:b448200849 "2"^^ . _:b448200336 . _:b448200337 . _:b448200338 . _:b448200509 "3"^^ . _:b448200339 . _:b448200848 "2"^^ . _:b448200348 . _:b448200349 . _:b448200643 . _:b448200350 . _:b448200508 "3"^^ . . _:b448200351 . _:b448200851 "2"^^ . _:b448200344 . _:b448200345 . _:b448200875 . _:b448200477 . _:b448200346 . . _:b448200511 "3"^^ . _:b448200347 . _:b448200850 "2"^^ . _:b448200324 . _:b448200325 . _:b448200326 . _:b448200544 . _:b9325946 . _:b448200861 "2"^^ . _:b448200327 . _:b448200510 "3"^^ . _:b448200320 . _:b448200321 . _:b448200322 . . _:b448200860 "2"^^ . _:b448200323 . _:b448200505 "3"^^ . _:b448200332 . _:b448200333 . _:b448200599 . _:b448200334 . _:b9325920 . _:b448200504 "3"^^ . _:b448200863 "2"^^ . _:b448200335 . _:b448200328 . . _:b448200329 . _:b448200330 . _:b448200507 "3"^^ . _:b448200862 "2"^^ . _:b448200331 . _:b448200436 . _:b448200437 . _:b448200698 . _:b448200438 . . _:b448200506 "3"^^ . _:b448200857 "2"^^ . _:b448200439 . _:b448200432 . _:b448200433 . _:b448200700 . _:b448200434 . _:b448200517 "3"^^ . _:b448200856 "2"^^ . _:b448200435 . _:b448200444 . _:b448200445 . _:b448200446 . _:b448200516 "3"^^ . _:b448200859 "2"^^ . _:b448200447 . _:b448200440 . _:b448200441 . _:b448200442 . _:b448200519 "3"^^ . _:b448200858 "2"^^ . _:b448200443 . _:b448200420 . _:b448200421 . _:b448200422 . _:b9325930 . _:b448200518 "3"^^ . _:b448200869 "2"^^ . _:b448200423 . _:b448200416 . _:b448200756 . _:b448200417 . _:b448200757 . _:b448200418 . _:b448200513 "3"^^ . _:b448200868 "2"^^ . . _:b448200419 . _:b448200758 . _:b448200759 . _:b448200428 . _:b448200752 . _:b448200429 . _:b448200753 . _:b9325928 . _:b448200430 . _:b448200512 "3"^^ . _:b448200883 . _:b448200871 "2"^^ . _:b448200431 . _:b448200754 . _:b448200755 . _:b9325925 "The binding assay methods for MHC class II molecules in general [42,>>43<<] as well as HLA-DP [44] and HLA-DQ [45] molecules have been described in detail previously." . _:b448200424 . _:b448200764 . _:b448200425 . _:b448200765 . _:b448200426 . _:b448200515 "3"^^ . _:b448200870 "2"^^ . _:b448200427 . _:b448200766 . _:b448200767 . _:b448200404 . . _:b448200405 . _:b448200760 . _:b448200761 . _:b448200406 . _:b448200514 "3"^^ . _:b448200865 "2"^^ . _:b448200407 . _:b448200618 . _:b448200762 . _:b448200763 . _:b448200400 . _:b448200740 . _:b448200401 . _:b448200741 . _:b448200402 . . _:b448200525 "3"^^ . _:b448200864 "2"^^ . _:b448200524 . _:b448200403 . _:b448200742 . _:b448200743 . _:b448200412 . _:b448200736 . _:b448200413 . _:b448200737 . _:b448200414 . _:b448200524 "3"^^ . _:b448200867 "2"^^ . _:b448200415 . _:b448200617 . _:b448200738 . _:b448200739 . _:b448200408 . _:b448200748 . _:b448200409 . _:b448200749 . _:b448200410 . _:b448200527 "3"^^ . _:b448200866 "2"^^ . _:b448200411 . _:b448200750 . _:b448200751 . _:b448200388 . _:b448200744 . _:b448200389 . . _:b448200745 . _:b448200390 . _:b448200526 "3"^^ . _:b448200877 "2"^^ . _:b448200391 . _:b448200746 . _:b448200747 . _:b448200384 . _:b448200724 . _:b448200385 . _:b448200725 . _:b448200386 . _:b448200793 . _:b448200568 . _:b448200876 "2"^^ . _:b448200387 . _:b448200726 . _:b448200521 "3"^^ . _:b448200727 . _:b448200396 . _:b448200720 . _:b448200397 . _:b448200721 . . _:b448200398 . _:b448200520 "3"^^ . _:b448200879 "2"^^ . . _:b448200399 . . _:b448200722 . _:b448200723 . _:b9325914 "on the matrices constructed by Sturniolo et al. [16] on which also the TEPITOPE predictions are based) were the top performing ones and were incorporated into the MHC class II binding prediction component of the IEDB analysis resource [>>21<<]. Here, we re-evaluated their performance on the new dataset." . _:b448200392 . _:b448200732 . . _:b448200393 . _:b448200733 . _:b448200394 . _:b448200510 . _:b448200523 "3"^^ . _:b448200878 "2"^^ . . _:b448200395 . _:b448200734 . _:b448200735 . . _:b448200728 . _:b448200729 . _:b448200522 "3"^^ . _:b448200873 "2"^^ . _:b448200730 . _:b448200731 . _:b448200708 . _:b448200533 "2"^^ . . _:b448200709 . _:b448200710 . _:b448200872 "2"^^ . _:b448200665 . _:b448200711 . _:b448200704 . _:b448200532 "2"^^ . _:b448200705 . _:b448200706 . _:b448200875 "2"^^ . _:b448200707 . _:b448200716 . . _:b448200717 . _:b448200535 "2"^^ . _:b448200718 . _:b448200874 "2"^^ . . _:b448200719 . _:b448200712 . _:b448200534 "2"^^ . _:b448200713 . _:b448200714 . _:b448200885 "2"^^ . _:b448200715 . _:b448200820 . _:b448200821 . . _:b448200529 "3"^^ . _:b448200884 "2"^^ . _:b448200822 . _:b448200585 . _:b448200536 . _:b448200823 . _:b448200816 . _:b448200817 . _:b448200528 "3"^^ . _:b448200887 "2"^^ . _:b448200818 . _:b448200819 . _:b448200828 . _:b448200829 . _:b448200531 "3"^^ . _:b448200886 "2"^^ . _:b448200830 . _:b448200831 . _:b448200824 . _:b448200825 . _:b448200530 "3"^^ . _:b448200881 "2"^^ . _:b448200826 . _:b448200827 . _:b448200804 . _:b448200541 "2"^^ . _:b448200805 . _:b448200806 . _:b448200880 "2"^^ . _:b448200807 . _:b448200800 . _:b448200540 "2"^^ . _:b448200801 . _:b448200523 . _:b448200883 "2"^^ . _:b448200802 . _:b448200803 . _:b448200812 . _:b448200543 "2"^^ . _:b448200813 . _:b448200814 . _:b448200882 "2"^^ . _:b448200815 . _:b448200808 . . _:b448200542 "2"^^ . _:b448200809 . _:b448200810 . . _:b448200390 . . _:b448200558 . _:b448200811 . _:b448200788 . _:b448200537 "2"^^ . _:b448200789 . _:b448200790 . _:b448200892 "2"^^ . _:b448200791 . _:b448200784 . _:b448200522 . _:b448200536 "2"^^ . _:b448200785 . _:b448200786 . _:b448200787 . _:b448200796 . _:b448200539 "2"^^ . _:b448200797 . _:b448200798 . _:b448200799 . _:b448200792 . _:b448200538 "2"^^ . _:b448200793 . _:b448200794 . _:b448200889 "2"^^ . _:b448200795 . _:b448200772 . _:b448200549 "2"^^ . _:b448200773 . _:b448200774 . _:b448200888 "2"^^ . _:b448200775 . _:b448200768 . _:b448200548 "2"^^ . _:b448200769 . _:b448200770 . _:b448200891 "2"^^ . _:b448200632 . _:b448200771 . _:b448200411 . _:b448200780 . _:b448200551 "2"^^ . . _:b448200781 . _:b448200782 . _:b448200890 "2"^^ . _:b448200783 . _:b448200776 . _:b448200550 "2"^^ . _:b448200777 . _:b448200778 . . _:b448200779 . _:b448200884 . _:b448200484 . _:b448200885 . _:b448200545 "2"^^ . _:b448200886 . _:b448200887 . _:b448200661 . _:b448200746 . . _:b448200880 . _:b448200544 "2"^^ . _:b448200881 . . _:b448200882 . _:b448200883 . _:b448200892 . _:b448200858 . _:b448200547 "2"^^ . _:b9325960 "used in the NetMHC algorithm [26,27] that was highly successful in predicting the binding specificity of HLA class I molecules [28,29] with the representation of peptide flanking residues and peptide length used in NetMHCIIpan method [>>19<<]. Both the NN-align and the combinatorial peptide library method were evaluated in terms of their prediction performance and ability to improve a consensus prediction approach." . _:b448200888 . _:b448200619 . _:b448200546 "2"^^ . _:b448200571 . _:b448200889 . _:b448200890 . . _:b448200891 . _:b448200849 . _:b448200868 . _:b448200557 "2"^^ . _:b448200869 . . _:b448200870 . _:b448200871 . _:b448200864 . _:b448200556 "2"^^ . _:b448200865 . _:b448200866 . _:b448200867 . _:b448200876 . _:b448200559 "2"^^ . _:b448200877 . _:b448200878 . . _:b448200879 . _:b448200872 . _:b448200558 "2"^^ . _:b448200873 . _:b448200874 . _:b448200875 . _:b448200397 . _:b448200852 . . _:b448200553 "2"^^ . _:b448200853 . _:b448200854 . . _:b448200855 . _:b448200848 . _:b448200552 "2"^^ . _:b448200849 . _:b448200850 . _:b448200851 . _:b448200860 . _:b448200555 "2"^^ . _:b448200861 . _:b448200862 . _:b448200863 . _:b448200856 . _:b448200464 . _:b448200857 . _:b448200554 "2"^^ . _:b448200858 . _:b448200859 . _:b448200836 . _:b448200565 "2"^^ . _:b448200837 . _:b448200838 . _:b448200649 . _:b448200839 . _:b9325965 . _:b448200832 . . _:b448200564 "2"^^ . _:b448200833 . _:b448200834 . _:b448200835 . _:b448200844 . _:b448200361 . _:b448200811 . _:b448200567 "2"^^ . _:b448200845 . _:b448200846 . _:b448200847 . _:b448200840 . _:b448200566 "2"^^ . _:b448200841 . _:b448200842 . _:b448200843 . _:b448200333 . _:b448200561 "2"^^ . _:b448200560 "2"^^ . _:b448200563 "2"^^ . _:b9325919 . _:b448200677 . _:b448200776 . _:b448200798 . _:b448200841 . _:b9325918 . _:b448200562 "2"^^ . . _:b448200573 "2"^^ . _:b448200572 "2"^^ . _:b448200575 "2"^^ . _:b448200574 "2"^^ . _:b448200569 "2"^^ . . _:b448200568 "2"^^ . _:b448200787 . . _:b448200571 "2"^^ . _:b448200570 "2"^^ . . _:b448200581 "2"^^ . _:b448200580 "2"^^ . _:b448200583 "2"^^ . . _:b448200859 . _:b448200582 "2"^^ . . _:b448200577 "2"^^ . _:b448200683 . . _:b448200788 . _:b448200576 "2"^^ . _:b448200656 . _:b448200579 "2"^^ . _:b448200578 "2"^^ . _:b448200589 "2"^^ . . . _:b448200588 "2"^^ . _:b448200326 . _:b448200697 . _:b448200591 "2"^^ . _:b448200590 "2"^^ . . _:b448200309 . . _:b448200585 "2"^^ . _:b448200584 "2"^^ . . _:b448200587 "2"^^ . _:b448200491 . _:b448200324 . . _:b448200586 "2"^^ . . _:b448200597 "2"^^ . _:b448200694 . _:b448200338 . _:b448200596 "2"^^ . _:b448200569 . _:b448200599 "2"^^ . _:b448200447 . _:b448200598 "2"^^ . _:b448200593 "2"^^ . _:b448200592 "2"^^ . _:b448200595 "2"^^ . _:b448200480 . . _:b448200382 . . _:b448200594 "2"^^ . _:b9325936 "HLA class II molecules are expressed by human professional antigen presenting cells (APCs) and can display peptides derived from exogenous antigens to CD4+ T cells [>>1<<]. The molecules are heterodimers consisting of an alpha chain and a beta chain encoded in one of three loci:" . . . _:b448200605 "2"^^ . . _:b448200604 "2"^^ . _:b448200638 . _:b448200607 "2"^^ . _:b448200606 "2"^^ . _:b448200864 . . _:b448200855 . _:b448200675 . _:b448200766 . _:b448200597 . _:b448200601 "2"^^ . . _:b448200761 . _:b448200600 "2"^^ . _:b448200596 . _:b448200378 . _:b448200603 "2"^^ . _:b448200602 "2"^^ . . _:b448200613 "2"^^ . . _:b448200612 "2"^^ . _:b448200615 "2"^^ . . _:b448200614 "2"^^ . _:b448200517 . _:b448200609 "2"^^ . _:b448200608 "2"^^ . _:b448200403 . . _:b448200850 . _:b448200611 "2"^^ . . . _:b448200610 "2"^^ . _:b448200385 . _:b448200621 "2"^^ . . _:b448200620 "2"^^ . _:b448200623 "2"^^ . . _:b448200622 "2"^^ . . _:b448200460 . _:b448200617 "2"^^ . _:b448200750 . _:b448200616 "2"^^ . . . _:b448200619 "2"^^ . . _:b448200362 . _:b448200520 . _:b448200618 "2"^^ . _:b448200701 . _:b448200566 . _:b448200629 "2"^^ . _:b448200628 "2"^^ . _:b448200631 "2"^^ . _:b448200630 "2"^^ . . _:b448200625 "2"^^ . . _:b448200465 . _:b448200624 "2"^^ . _:b448200472 . _:b448200627 "2"^^ . _:b448200842 . _:b448200626 "2"^^ . . _:b448200350 . _:b9325929 . _:b448200637 "2"^^ . . . _:b448200636 "2"^^ . . . _:b448200639 "2"^^ . _:b448200678 . _:b448200638 "2"^^ . _:b448200813 . _:b448200866 . _:b448200633 "2"^^ . _:b448200632 "2"^^ . . . _:b448200635 "2"^^ . . _:b448200634 "2"^^ . _:b448200645 "2"^^ . . _:b448200644 "2"^^ . _:b448200647 "2"^^ . . _:b448200646 "2"^^ . _:b448200398 . _:b448200641 "2"^^ . _:b448200640 "2"^^ . . _:b448200643 "2"^^ . _:b448200863 . _:b448200642 "2"^^ . _:b448200653 "2"^^ . . . _:b448200652 "2"^^ . _:b448200317 . . _:b448200655 "2"^^ . . _:b448200584 . _:b448200368 . _:b448200654 "2"^^ . _:b448200630 . _:b448200649 "2"^^ . _:b448200744 . _:b448200648 "2"^^ . _:b448200651 "2"^^ . _:b448200650 "2"^^ . . . _:b448200345 . _:b448200661 "2"^^ . . _:b448200622 . _:b448200660 "2"^^ . _:b448200663 "2"^^ . . _:b448200332 . _:b448200662 "2"^^ . _:b448200306 . _:b448200814 . _:b448200657 "2"^^ . _:b448200656 "2"^^ . _:b448200659 "2"^^ . . _:b448200658 "2"^^ . _:b448200417 . . _:b448200669 "2"^^ . _:b448200409 . _:b448200668 "2"^^ . _:b448200671 "2"^^ . _:b448200498 . _:b448200670 "2"^^ . . . _:b448200665 "2"^^ . _:b448200664 "2"^^ . _:b448200667 "2"^^ . _:b448200351 . _:b448200666 "2"^^ . _:b448200603 . _:b448200677 "2"^^ . _:b448200676 "2"^^ . . _:b448200679 "2"^^ . _:b448200678 "2"^^ . . _:b448200673 "2"^^ . _:b448200672 "2"^^ . _:b448200675 "2"^^ . _:b448200401 . _:b448200778 . _:b448200674 "2"^^ . . _:b448200381 . _:b448200685 "2"^^ . . _:b448200684 "2"^^ . . _:b448200687 "2"^^ . _:b9325951 . . _:b448200686 "2"^^ . _:b448200681 "2"^^ . . _:b448200680 "2"^^ . . . _:b448200683 "2"^^ . _:b448200349 . _:b448200682 "2"^^ . _:b9325976 . _:b448200693 "2"^^ . _:b448200692 "2"^^ . _:b448200695 "2"^^ . _:b9325954 . _:b448200694 "2"^^ . . _:b448200689 "2"^^ . _:b448200573 . _:b448200688 "2"^^ . _:b448200667 . . _:b448200691 "2"^^ . . _:b448200646 . _:b448200690 "2"^^ . . . _:b448200701 "2"^^ . _:b448200700 "2"^^ . _:b9325939 "The DR locus can encode two beta chains DRB1 and DRB3-5 which are in linkage disequilibrium [>>4<<]. The genes encoding class II molecules are highly polymorphic, as evidenced by the IMGT/HLA database [5] which lists 1,190 known sequences of HLA class II alleles for HLA-DR, HLA-DP and HLA-DQ molecules (Table 1). Both alpha and beta" . . _:b448200703 "2"^^ . _:b448200580 . . . _:b448200702 "2"^^ . . _:b448200697 "2"^^ . . . _:b448200696 "2"^^ . . _:b448200664 . _:b448200600 . . _:b448200699 "2"^^ . . _:b448200698 "2"^^ . _:b9325969 . _:b448200812 . . _:b9325913 . _:b448200521 . _:b448200516 . . . . _:b448200432 . _:b448200364 . _:b448200860 . _:b9325975 . . . . . _:b448200644 . . _:b9325960 . . _:b448200406 . _:b9325966 . _:b448200556 . . _:b448200425 . _:b448200879 . . _:b448200829 . _:b448200459 . _:b448200413 . _:b448200780 . _:b448200363 . _:b448200512 . _:b9325926 "The binding assay methods for MHC class II molecules in general [42,43] as well as HLA-DP [>>44<<] and HLA-DQ [45] molecules have been described in detail previously." . . . _:b448200760 . . _:b448200329 . . _:b9325916 "The previously reported prediction performance data taken from [>>30<<] is also shown in Table 3 under the \"old\" columns. Compared to the average evaluation results reported previously, ARB (0.784 vs." . . . . . . _:b448200809 . _:b448200457 . _:b448200636 . _:b9325951 . . _:b448200370 . . . _:b448200658 . _:b448200810 . . _:b448200607 . . . _:b448200792 . _:b9325938 . _:b9325964 "Lack of data for these alleles was identified in previous studies as one of the challenges facing HLA class II binding predictions [>>30<<,31]. The significant increase (i.e. over 40%) in the number of allelic variants results in a > 99% population coverage which could be very valuable for the development of T-cell epitope based vaccine. This dataset will also be useful in" . . _:b448200560 . . _:b448200319 . . . _:b448200624 . . . _:b448200820 . . _:b448200380 . _:b448200628 . . . . . . _:b448200471 . _:b448200453 . _:b448200728 . _:b448200452 . _:b448200455 . . _:b448200753 . _:b448200454 . _:b448200449 . . _:b448200448 . _:b448200415 . _:b448200564 . _:b448200451 . _:b448200450 . _:b448200461 . . _:b448200460 . _:b448200557 . _:b448200575 . _:b448200463 . . _:b448200645 . _:b448200462 . . _:b448200393 . _:b448200457 . _:b448200737 . _:b448200456 . _:b448200601 . _:b9325921 . _:b448200839 . _:b448200459 . _:b448200458 . . _:b448200469 . . _:b9325945 . _:b448200518 . _:b448200468 . _:b448200561 . _:b448200714 . . _:b448200471 . _:b9325931 "The cutoff of 1000 nM was chosen for its biological relevance as a previous study showed that a cutoff of 1000 nM captured near 97% DR-restricted epitopes [>>50<<]. For each peptide in a partition, we first determined its similarity with the rest of peptides in the dataset and the number of peptides sharing similarity with each peptide (Nsimilarity) was recorded. We then sorted the peptides" . . . _:b448200470 . _:b448200795 . . . _:b448200465 . _:b448200464 . . _:b448200825 . . _:b448200467 . _:b448200727 . _:b9325963 "The binding dataset made available here is about four-fold larger than the one in our previous report [>>30<<]. The increased number of peptides per allele resulted in a significantly improved performance of machine learning methods, ARB and SMM-align. This reinforces the idea that the prediction performance of a machine learning method is" . _:b448200466 . _:b448200477 . _:b9325935 "background" . . _:b448200476 . . _:b448200479 . _:b448200478 . _:b448200699 . _:b448200473 . _:b448200532 . _:b448200723 . _:b448200472 . _:b448200475 . . . . _:b448200474 . _:b448200485 . . _:b448200484 . _:b448200487 . . _:b448200486 . _:b448200686 . . _:b448200660 . _:b448200481 . . _:b448200480 . _:b448200662 . _:b448200483 . _:b448200845 . _:b448200796 . _:b9325944 "Several databases have been established to document the results of such experiments including Antijen [>>9<<], MHCBN [10], MHCPEP [11], FIMM [12], SYFPEITHI [13] and the Immune Epitope Database (IEDB) [14,15]." . _:b448200482 . . _:b448200499 . _:b448200493 . _:b448200475 . _:b448200492 . _:b448200495 . _:b448200494 . _:b448200489 . _:b448200488 . _:b448200887 . _:b448200491 . _:b448200490 . _:b448200501 . "PMC0" . . _:b448200410 . _:b448200500 . _:b448200483 . _:b448200797 . _:b448200503 . _:b448200502 . _:b448200497 . _:b448200312 . . _:b448200496 . _:b448200499 . _:b448200498 . . _:b448200502 . _:b448200509 . _:b448200508 . _:b9325927 "The binding assay methods for MHC class II molecules in general [42,43] as well as HLA-DP [44] and HLA-DQ [>>45<<] molecules have been described in detail previously." . _:b448200511 . _:b448200510 . _:b448200505 . _:b448200504 . _:b448200507 . _:b448200654 . _:b448200506 . _:b448200517 . _:b448200516 . _:b448200659 . . _:b448200519 . _:b448200518 . _:b448200356 . . _:b448200513 . . _:b448200512 . _:b448200682 . _:b448200734 . _:b448200621 . _:b448200515 . . _:b448200514 . _:b448200525 . _:b448200458 . _:b448200524 . _:b448200655 . _:b448200527 . _:b448200526 . . _:b448200521 . _:b448200520 . _:b448200684 . _:b448200523 . _:b448200434 . _:b448200522 . _:b448200533 . . _:b448200532 . _:b448200528 . _:b448200421 . _:b448200547 . . _:b448200803 . _:b448200535 . _:b448200668 . _:b448200791 . _:b448200534 . . _:b448200529 . . _:b448200304 . _:b448200511 . _:b448200528 . _:b448200716 . _:b9325973 "One of the challenges for evaluating the MHC class II binding prediction performances is how to deal with the presence of homologous peptides in the available data [>>32<<]. One concern is that peptides in the testing set for which a homolog is present in the training data may lead to artificially high prediction performances." . _:b448200555 . _:b448200531 . _:b448200530 . _:b9325921 "relevant this finding is in practice, specifically as the inclusion of single residue substitutions can contain particularly useful information demonstrated by the fact that this is how the MHC binding motifs were originally defined [>>33<<]." . _:b448200541 . _:b448200540 . . _:b448200543 . _:b448200354 . _:b448200542 . _:b448200501 . _:b448200533 . _:b448200537 . . _:b448200441 . _:b448200536 . . _:b448200539 . . _:b9325950 . . _:b9325932 . _:b448200823 . _:b448200538 . _:b448200549 . _:b9325951 "It is possible to develop binding prediction methods for HLA molecules for which no experimental data are available by extrapolating what is known for related molecules [>>16<<-19]. However, the quality of these extrapolations decreases for molecules that are very different from the experimentally characterized ones, and completely ab initio predictions have not been successful [20]. It is therefore a major gap" . _:b448200639 . _:b448200548 . _:b448200635 . _:b448200394 . . _:b448200551 . _:b448200805 . _:b448200550 . _:b448200545 . _:b9325962 . _:b448200544 . _:b9325932 "The forward step-wise Hobohm 1 algorithm [>>51<<] consisting of the following three steps was next applied to generate a similarity reduced:" . _:b9325942 "HLA class II peptide ligands that are recognized by T cells and trigger an immune response are referred to as immune epitopes [>>7<<]. Identifying such epitopes can help detect and modulate immune responses in infectious diseases, allergy, autoimmune diseases and cancer." . _:b448200311 . _:b448200387 . _:b448200547 . _:b448200546 . . _:b448200557 . . . _:b448200758 . _:b448200556 . . _:b448200559 . _:b448200558 . _:b9325948 "Several databases have been established to document the results of such experiments including Antijen [9], MHCBN [10], MHCPEP [11], FIMM [12], SYFPEITHI [>>13<<] and the Immune Epitope Database (IEDB) [14,15]." . _:b9325969 "Data from such libraries have successfully been used to predict proteasomal cleavage [22], TAP transport [23] and MHC class I binding [>>24<<]. The performance of the libraries for class II predictions was comparable to that of PROPRED, and in general inferior to the machine learning approaches. The main value of the combinatorial library approach lies in its experimental" . _:b448200553 . _:b448200526 . _:b448200552 . _:b448200847 . _:b448200555 . _:b448200594 . _:b448200747 . _:b448200554 . _:b448200565 . . _:b448200564 . _:b448200567 . _:b448200323 . . _:b448200566 . . _:b448200770 . _:b9325971 "This repeats the dominating performance of the related NetMHC prediction methods in a number of recent MHC class I prediction benchmarks [28,>>29<<,38]." . _:b448200561 . . _:b448200560 . _:b448200878 . _:b448200563 . _:b448200562 . . _:b448200573 . _:b448200572 . _:b448200575 . . _:b448200416 . _:b448200574 . _:b448200876 . . . _:b448200569 . . _:b448200568 . _:b448200571 . _:b448200614 . _:b448200657 . _:b448200570 . _:b9325952 "However, the quality of these extrapolations decreases for molecules that are very different from the experimentally characterized ones, and completely ab initio predictions have not been successful [>>20<<]. It is therefore a major gap in knowledge that little binding data are available for HLA DP and DQ molecules, which are more difficult to work with experimentally, but are equally relevant as HLA DR molecules. Resulting from this lack of" . . . _:b448200581 . . _:b448200580 . . _:b448200583 . . _:b448200582 . . _:b448200577 . _:b448200576 . _:b448200579 . _:b448200578 . _:b448200589 . _:b448200705 . _:b448200588 . . . _:b448200591 . _:b448200590 . . _:b448200891 . _:b448200768 . _:b9325955 "time and cost effective, as the same panel of peptide libraries can be scanned for all HLA molecules of interest, and has been applied successfully for multiple applications [22-24], The second method we newly implemented was NN-align [>>25<<]. This neural network based approach combines the peptide sequence representation used in the NetMHC algorithm [26,27] that was highly successful in predicting the binding specificity of HLA class I molecules [28,29] with the" . _:b448200585 . _:b448200584 . _:b448200587 . . _:b9325923 . _:b448200586 . _:b448200597 . _:b9325936 . _:b448200519 . _:b448200596 . . _:b448200599 . . _:b448200598 . _:b9325951 . . . _:b448200593 . _:b448200688 . _:b448200588 . . _:b448200592 . _:b448200595 . . . _:b448200594 . . . _:b448200605 . _:b9325929 "Several previous studies have proposed measurements to determine peptide similarity [32,>>46<<-49]. Here we adopted the similarity measure described by El-Manzalawy et al. [32]." . _:b448200789 . _:b448200721 . . _:b448200604 . _:b448200607 . _:b448200606 . _:b448200344 . _:b9325968 . _:b448200601 . _:b448200786 . _:b448200600 . . _:b448200603 . . _:b448200570 . _:b448200487 . _:b448200602 . _:b448200613 . _:b448200495 . . _:b448200835 . . . _:b448200612 . _:b448200615 . _:b448200614 . _:b448200826 . _:b448200609 . _:b448200608 . _:b448200545 . _:b448200611 . . _:b448200836 . _:b448200610 . _:b448200693 . . _:b448200621 . _:b448200620 . _:b9325950 "Several databases have been established to document the results of such experiments including Antijen [9], MHCBN [10], MHCPEP [11], FIMM [12], SYFPEITHI [13] and the Immune Epitope Database (IEDB) [14,>>15<<]. IEDB currently documents 12,577 peptides tested for binding to one of more of 158 MHC class II allelic variants of which 114 are human (HLA). It is possible to develop binding prediction methods for HLA molecules for which no" . _:b448200623 . _:b9325956 . _:b448200622 . . _:b448200617 . . . _:b448200616 . _:b9325954 . _:b9325908 . _:b448200619 . _:b448200756 . . _:b9325909 . _:b448200618 . _:b9325910 . _:b448200629 . _:b9325911 . _:b9325934 . _:b448200628 . _:b448200529 . _:b448200631 . . _:b448200552 . _:b448200775 . _:b448200630 . . _:b448200625 . . _:b448200476 . _:b448200624 . _:b448200593 . _:b9325916 . _:b448200627 . _:b9325917 . _:b448200626 . _:b9325918 . . _:b448200637 . _:b9325919 . _:b448200636 . . _:b448200724 . _:b448200598 . _:b9325912 . _:b448200639 . _:b9325913 . . _:b448200638 . _:b9325914 . . _:b448200633 . _:b448200399 . _:b9325915 . _:b448200632 . _:b9325924 . _:b448200635 . _:b9325967 "Data from such libraries have successfully been used to predict proteasomal cleavage [>>22<<], TAP transport [23] and MHC class I binding [24]." . _:b9325925 . _:b448200634 . . _:b9325956 "This neural network based approach combines the peptide sequence representation used in the NetMHC algorithm [>>26<<,27] that was highly successful in predicting the binding specificity of HLA class I molecules [28,29] with the representation of peptide flanking residues and peptide length used in NetMHCIIpan method [19]." . _:b9325926 . _:b448200645 . _:b448200366 . _:b9325927 . _:b448200644 . . _:b448200389 . _:b9325920 . _:b448200647 . _:b9325972 "This repeats the dominating performance of the related NetMHC prediction methods in a number of recent MHC class I prediction benchmarks [28,29,>>38<<]." . _:b9325921 . _:b448200646 . _:b9325922 . _:b448200641 . _:b448200479 . _:b448200799 . _:b9325923 . _:b448200640 . . _:b9325932 . _:b448200643 . _:b9325933 . _:b448200642 . _:b9325934 . _:b448200653 . _:b9325935 . _:b448200652 . . _:b9325928 . _:b448200321 . _:b448200655 . _:b9325929 . _:b448200654 . _:b9325947 . _:b448200541 . _:b9325930 . _:b448200649 . . _:b9325931 . . _:b448200648 . . _:b9325940 . _:b448200496 . _:b448200651 . _:b9325941 . _:b448200650 . _:b448200433 . _:b9325942 . _:b448200661 . _:b448200431 . _:b9325943 . . _:b448200660 . _:b448200815 . _:b448200490 . _:b9325936 . _:b448200314 . _:b448200663 . _:b9325937 . _:b448200662 . . . _:b9325976 "We plan to systematically examine how to best construct consensus predictions for MHC binding in the future, building on work done by us and others in the past [30,39,>>40<<]." . _:b9325938 . _:b448200657 . _:b448200689 . _:b9325939 . _:b448200656 . _:b448200679 . _:b9325948 . _:b448200659 . _:b448200325 . _:b9325949 . _:b448200658 . _:b448200303 . _:b9325950 . _:b448200669 . _:b448200493 . . _:b9325951 . _:b448200668 . _:b9325962 "Computational algorithms to predict epitope candidates have become an essential tool for genomic screens of pathogens for T cell response targets [>>34<<-37]. The majority of these algorithms rely on experimental binding affinities to generate predictive models." . _:b9325944 . _:b448200671 . . . . _:b9325945 . _:b448200670 . _:b9325946 . _:b448200665 . _:b448200515 . _:b9325924 "The combinatorial libraries were synthesized as previously described [24,>>41<<]. Peptides in each library are 13-mers with Alanine residues in positions 1, 2, 12 and 13." . _:b9325947 . _:b448200664 . . . _:b9325956 . _:b448200667 . _:b9325957 . _:b448200666 . _:b9325958 . _:b448200677 . _:b448200871 . _:b9325959 . _:b448200676 . _:b9325952 . . _:b448200679 . _:b448200692 . _:b448200386 . _:b9325953 . _:b448200678 . _:b9325954 . _:b448200673 . _:b9325955 . _:b448200672 . _:b9325964 . _:b448200675 . _:b9325965 . _:b448200674 . _:b9325966 . _:b448200685 . _:b448200328 . _:b9325967 . . _:b448200684 . _:b9325960 . _:b448200687 . . _:b9325961 . _:b448200686 . _:b448200467 . _:b9325962 . _:b448200741 . _:b448200681 . _:b448200559 . _:b9325963 . . _:b448200680 . _:b9325972 . _:b448200683 . . _:b9325973 . _:b448200682 . _:b448200605 . _:b9325974 . _:b448200693 . _:b9325975 . _:b448200692 . . _:b9325968 . _:b448200695 . _:b9325910 "Secondly, for several molecules, relatively few data points existed, in spite of the fact that we and others [>>30<<,31] have shown that several hundred data points are desirable to derive accurate predictive algorithms." . _:b9325969 . _:b448200430 . _:b448200694 . _:b9325970 . _:b448200689 . _:b9325971 . _:b448200688 . _:b448200691 . _:b448200690 . _:b448200701 . _:b448200700 . . _:b9325976 . _:b448200703 . _:b448200702 . . _:b448200697 . _:b448200696 . _:b448200892 . _:b448200699 . . _:b448200698 . _:b448200549 . . . . . _:b448200637 . . _:b448200650 . _:b448200771 . . _:b448200709 . _:b448200611 . _:b448200708 . _:b448200862 . _:b448200711 . _:b448200437 . _:b448200710 . . . _:b448200705 . . _:b448200685 . _:b448200704 . _:b448200707 . _:b448200706 . _:b448200717 . _:b448200716 . _:b448200455 . _:b448200719 . _:b448200718 . _:b448200713 . _:b448200712 . _:b448200715 . _:b448200714 . _:b448200652 . _:b448200537 . _:b448200725 . . . _:b448200877 . _:b448200724 . _:b448200727 . _:b448200726 . . _:b448200589 . _:b448200721 . _:b9325949 . _:b448200720 . _:b448200723 . . . _:b448200451 . _:b448200651 . _:b448200838 . _:b448200722 . _:b448200733 . _:b448200367 . . _:b448200732 . _:b448200754 . _:b448200735 . _:b448200734 . _:b448200729 . _:b448200728 . . _:b448200305 . _:b448200731 . . _:b448200730 . _:b448200817 . _:b448200507 . _:b448200741 . _:b448200740 . _:b448200438 . _:b9325942 . _:b448200743 . . . _:b448200742 . . _:b448200737 . _:b448200384 . _:b448200736 . _:b9325937 "The molecules are heterodimers consisting of an alpha chain and a beta chain encoded in one of three loci: HLA DR, DP and DQ [>>2<<,3]. The DR locus can encode two beta chains DRB1 and DRB3-5 which are in linkage disequilibrium [4]. The genes encoding class II molecules are highly polymorphic, as evidenced by the IMGT/HLA database [5] which lists 1,190 known sequences" . _:b448200739 . _:b448200313 . . _:b448200738 . _:b448200749 . . _:b448200748 . _:b448200751 . _:b448200550 . _:b448200750 . . _:b448200745 . _:b448200377 . . _:b448200744 . _:b448200642 . _:b448200747 . . _:b448200746 . _:b448200757 . . _:b448200806 . _:b448200756 . . _:b448200759 . . _:b448200758 . _:b448200725 . _:b448200753 . . _:b448200623 . _:b9325947 "Several databases have been established to document the results of such experiments including Antijen [9], MHCBN [10], MHCPEP [11], FIMM [>>12<<], SYFPEITHI [13] and the Immune Epitope Database (IEDB) [14,15]." . _:b448200752 . _:b448200755 . _:b448200754 . _:b448200765 . . _:b448200764 . . _:b448200767 . _:b448200766 . . _:b448200761 . _:b448200760 . . _:b448200604 . _:b448200763 . . _:b448200762 . _:b448200773 . _:b448200531 . . _:b448200772 . . _:b448200775 . _:b448200774 . _:b448200602 . _:b448200769 . _:b448200768 . _:b448200540 . _:b448200771 . _:b448200830 . _:b448200612 . _:b448200770 . _:b448200316 . _:b448200781 . _:b448200780 . . _:b448200783 . _:b448200353 . . _:b448200782 . _:b448200720 . _:b448200777 . _:b448200735 . _:b448200776 . . _:b448200779 . _:b448200831 . _:b448200778 . _:b448200563 . _:b448200789 . . . _:b448200788 . _:b448200708 . _:b448200791 . _:b448200790 . _:b448200785 . _:b448200784 . . _:b448200787 . _:b448200851 . _:b448200436 . . _:b448200786 . . _:b448200886 . . . _:b448200797 . . _:b448200796 . _:b448200799 . _:b448200798 . _:b448200793 . . _:b448200792 . _:b448200795 . _:b448200794 . _:b448200373 . _:b448200805 . _:b448200804 . _:b448200303 . _:b448200807 . _:b448200513 . _:b448200806 . _:b448200357 . _:b448200801 . _:b448200592 . _:b448200800 . _:b448200739 . _:b9325922 "methods" . _:b448200803 . _:b9325931 . _:b448200802 . _:b448200309 . _:b448200374 . _:b9325939 . _:b448200813 . _:b448200308 . _:b448200812 . _:b448200311 . _:b448200824 . . _:b448200815 . _:b9325948 . _:b448200730 . _:b448200310 . _:b448200308 . _:b448200702 . _:b448200814 . _:b9325972 . _:b448200305 . _:b448200310 . _:b9325908 "results" . _:b448200809 . _:b448200304 . _:b448200808 . _:b448200307 . _:b448200811 . _:b9325965 "Lack of data for these alleles was identified in previous studies as one of the challenges facing HLA class II binding predictions [30,>>31<<]. The significant increase (i.e. over 40%) in the number of allelic variants results in a > 99% population coverage which could be very valuable for the development of T-cell epitope based vaccine. This dataset will also be useful in" . _:b448200306 . _:b448200810 . . _:b448200317 . _:b448200821 . _:b448200316 . _:b448200640 . _:b448200336 . _:b448200820 . . _:b448200319 . . _:b448200823 . _:b448200318 . . _:b448200822 . _:b448200313 . _:b448200696 . _:b448200817 . _:b448200312 . . _:b448200816 . _:b448200315 . _:b448200819 . _:b448200314 . _:b9325940 "The genes encoding class II molecules are highly polymorphic, as evidenced by the IMGT/HLA database [>>5<<] which lists 1,190 known sequences of HLA class II alleles for HLA-DR, HLA-DP and HLA-DQ molecules (Table 1)." . _:b448200818 . . _:b448200325 . . _:b448200829 . _:b448200324 . _:b448200828 . _:b448200327 . _:b448200414 . _:b448200831 . _:b9325961 _:b9325962 . _:b448200326 . _:b9325961 _:b9325963 . _:b9325961 _:b9325964 . _:b448200774 . _:b9325961 _:b9325965 . _:b448200830 . _:b9325961 _:b9325966 . _:b448200321 . . _:b9325961 _:b9325967 . _:b9325961 _:b9325968 . _:b9325961 _:b9325969 . _:b448200825 . _:b448200339 . . _:b9325961 _:b9325970 . _:b448200320 . _:b9325961 _:b9325971 . _:b9325961 _:b9325972 . . _:b9325961 _:b9325973 . _:b448200824 . . _:b9325961 _:b9325974 . _:b448200323 . _:b9325961 _:b9325975 . _:b9325961 _:b9325976 . . _:b448200827 . _:b448200322 . . _:b448200530 . _:b448200365 . _:b448200826 . _:b448200333 . _:b448200837 . _:b448200332 . _:b448200836 . _:b448200335 . _:b448200844 . _:b448200839 . _:b448200334 . _:b448200838 . _:b448200329 . _:b448200833 . _:b448200328 . _:b9325943 "findings from peptide binding assays, sequencing of naturally presented HLA ligands, and three dimensional structures of HLA peptide complexes solved by X-ray crystallography (for a review on MHC class II prediction algorithms see [>>8<<] and references herein). Several databases have been established to document the results of such experiments including Antijen [9], MHCBN [10], MHCPEP [11], FIMM [12], SYFPEITHI [13] and the Immune Epitope Database (IEDB) [14,15]." . _:b448200832 . . _:b448200331 . . . _:b448200835 . _:b448200330 . _:b448200834 . _:b448200341 . . _:b448200845 . _:b448200340 . _:b448200488 . _:b448200764 . _:b448200844 . _:b448200343 . _:b448200691 . _:b448200847 . _:b448200342 . _:b448200846 . _:b448200337 . _:b448200841 . _:b448200336 . _:b448200423 . _:b448200840 . _:b448200497 . _:b448200339 . _:b448200843 . _:b448200338 . _:b448200842 . _:b448200349 . _:b448200706 . _:b448200853 . _:b448200348 . . _:b448200852 . _:b448200351 . . _:b448200855 . _:b448200350 . _:b448200854 . _:b9325915 "The old AUC values were taken from previous evaluation [>>30<<]." . _:b448200345 . _:b448200849 . . _:b448200344 . . . _:b448200848 . _:b448200347 . . _:b448200851 . _:b448200452 . _:b448200346 . . _:b448200850 . _:b448200357 . . _:b448200861 . _:b448200356 . _:b448200860 . _:b448200359 . . . _:b448200863 . . _:b448200358 . _:b448200862 . _:b448200353 . . _:b448200857 . _:b9325968 "Data from such libraries have successfully been used to predict proteasomal cleavage [22], TAP transport [>>23<<] and MHC class I binding [24]." . _:b448200352 . _:b448200371 . _:b448200669 . . _:b448200807 . _:b448200856 . _:b448200355 . _:b448200808 . _:b448200859 . _:b448200354 . _:b448200858 . _:b448200365 . . _:b448200869 . _:b448200364 . . . _:b448200868 . _:b448200367 . . _:b448200871 . _:b448200366 . . _:b448200481 . _:b448200870 . _:b448200361 . _:b448200865 . _:b448200360 . _:b448200864 . _:b448200363 . . . _:b448200711 . _:b448200867 . _:b448200307 . _:b448200362 . _:b448200866 . _:b448200373 . _:b448200690 . _:b9325922 _:b9325932 . _:b9325966 "This dataset will also be useful in improving pan-like approaches that take advantage of binding pocket similarities among different MHC molecules to generate binding predictors for allelic variants without binding data [>>19<<]." . _:b9325922 _:b9325933 . _:b9325922 _:b9325934 . _:b448200877 . . _:b448200372 . _:b9325922 _:b9325928 . _:b9325922 _:b9325929 . _:b9325922 _:b9325930 . _:b448200876 . _:b448200375 . _:b9325922 _:b9325931 . _:b9325922 _:b9325924 . _:b9325922 _:b9325925 . _:b9325922 _:b9325926 . _:b448200879 . _:b448200374 . _:b9325922 _:b9325927 . . _:b448200878 . _:b448200369 . _:b9325922 _:b9325923 . _:b448200819 . _:b448200873 . _:b448200368 . . _:b448200872 . _:b448200371 . _:b448200743 . _:b448200832 . _:b448200875 . _:b448200370 . . _:b448200874 . _:b448200381 . . _:b9325935 _:b9325944 . _:b9325935 _:b9325945 . _:b9325935 _:b9325946 . _:b448200885 . _:b448200380 . _:b9325935 _:b9325947 . . _:b9325935 _:b9325948 . _:b448200674 . _:b9325935 _:b9325949 . _:b9325935 _:b9325950 . _:b448200884 . _:b448200383 . _:b9325935 _:b9325951 . _:b9325935 _:b9325936 . _:b9325935 _:b9325937 . _:b9325935 _:b9325938 . _:b448200887 . _:b448200382 . _:b9325935 _:b9325939 . _:b9325935 _:b9325940 . _:b9325935 _:b9325941 . _:b9325935 _:b9325942 . _:b448200886 . _:b448200377 . _:b9325935 _:b9325943 . _:b448200881 . _:b448200376 . . . _:b448200880 . _:b448200379 . . _:b448200748 . . . _:b448200883 . _:b448200378 . _:b448200882 . _:b448200389 . _:b448200388 . . _:b448200504 . _:b448200892 . _:b448200391 . _:b448200390 . _:b448200620 . _:b448200385 . _:b448200889 . _:b448200384 . . _:b448200888 . _:b448200387 . . _:b448200891 . _:b448200386 . . _:b448200379 . _:b448200890 . _:b448200397 . . _:b448200396 . _:b448200399 . _:b448200784 . _:b448200503 . . _:b448200398 . . . _:b448200393 . _:b448200486 . _:b448200392 . _:b448200395 . _:b448200833 . . _:b448200394 . _:b448200405 . . _:b448200404 . . . _:b448200407 . _:b448200406 . _:b448200401 . . _:b448200372 . _:b448200400 . _:b448200852 . . _:b448200403 . _:b448200527 . _:b448200889 . _:b448200402 . _:b448200413 . . _:b448200583 . _:b448200412 . . _:b448200415 . _:b448200414 . _:b448200383 . _:b448200409 . _:b448200395 . _:b448200408 . _:b448200781 . _:b448200411 . _:b448200767 . _:b448200410 . _:b448200359 . _:b448200421 . _:b448200553 . _:b448200420 . _:b448200423 . _:b448200422 . . _:b448200417 . _:b9325935 _:b9325960 . _:b448200416 . _:b9325908 _:b9325920 . _:b9325908 _:b9325921 . _:b448200419 . _:b9325935 _:b9325952 . _:b9325935 _:b9325953 . _:b9325935 _:b9325954 . _:b448200418 . _:b9325935 _:b9325955 . . _:b9325935 _:b9325956 . _:b9325935 _:b9325957 . _:b9325935 _:b9325958 . _:b448200346 . _:b448200429 . _:b9325935 _:b9325959 . _:b9325908 _:b9325909 . _:b9325908 _:b9325910 . _:b448200428 . _:b9325908 _:b9325911 . _:b448200431 . . _:b9325908 _:b9325916 . _:b9325908 _:b9325917 . . . _:b9325908 _:b9325918 . _:b448200430 . _:b9325908 _:b9325919 .