Table 5.

Read filtering for differences in library preparation

		Preprocessing
Platform	Name	Raw	Filtered	Mapped	Deduplicated
DNBseq™	ERR1831362	48 148 821	45 828 900	44 015 469	32 497 907
	ERR1831363	29 782 959	28 352 783	27 237 354	21 102 998
	ERR1831364	54 940 056	52 466 727	50 421 993	36 693 200
	ERR1831365	36 073 210	34 329 221	32 932 438	24 745 818
	ERR1831366	43 664 065	41 782 043	40 108 615	29 250 048
	ERR1831367	55 025 946	52 410 892	50 302 197	36 153 502
	ERR1831368	53 296 161	50 698 009	48 688 107	34 475 744
	ERR1831369	65 455 754	62 428 710	59 984 288	41 622 545
	ERR1831370	29 774 053	28 307 062	27 164 676	20 606 457
HiSeq	SRR1261168	134 921 154	104 132 308	101 903 998	70 430 950
	SRR1261170	72 897 482	33 367 214	32 597 422	27 498 074
	SRR950078	100 387 010	77 761 236	73 979 293	50 505 406
	SRR950080	91 781 477	69 875 633	66 896 955	49 310 706
	SRR950084	125 083 194	93 367 543	89 192 069	61 653 799
DNBseq™	Total	416 161 025	396 604 347	380 855 137	277 148 219
	% Removed		4.70%	3.97%	27.23%
HiSeq	Total	525 070 317	378 503 934	364 569 737	259 398 935
	% Removed		27.91%	3.68%	28.85%

Here we show reads remaining after each preprocessing step. The columns indicate read counts after SOAPnuke filtering (Filtered), aligning to GRCh38 with HISAT2 (Mapped), and PCR deduplication with Picard Tools (Deduplicated)