Table 1.

iRF model summary and metrics

Model species	Source dataset	# of sgRNAs	Feature det	# of Features	Description	Test dpecies	R2	Pearson correlation
E.coli	Guo et al. 2016	40 468 [32 374 train]	Raw	5	summary values of sgRNA sequence including GC content, Tm, MFE, gene density and distance to PAM	E. coli	0.0406861	0.2007612
			Onehot	5911	binary positional encoding of 20bp sgRNA nucleotide sequence	E. coli	0.26004285	0.4914184
			QCT	316	quantitative metrics for H-bond and HL-gap based on positional nucleotide sequence	E. coli	0.24183122	0.4918057
			Raw.Onehot	6916	Raw + Onehot	E. coli	0.26028286	0.4931724
			Raw.QCT	312	Raw + QCT	E. coli	0.24177446	0.4939777
			Onehot.QCT	6227	Onehot + QCT	E. coli	0.24905183	0.500817
			Full Matrix	6232	Raw + Onehot + QCT	E. coli	0.24906667	0.5019173
						H. sapien	0.00429969	0.06557198
			Top 5	5	Based on the full feature matrix iRF model run with E.coli data, the top feature importance scores were utilized to generate new iRF modesl with 5,10,20,40100200500 and 1000 features.	E. coli	0.11240746	0.3436711
			Top 10	10		E. coli	0.15779734	0.4019815
			Top 20	20		E. coli	0.2017236	0.4458406
			Top 50	50		E. coli	0.24529071	0.4903894
			Top 100	100		E. coli	0.25119027	0.4967809
			Top 200	200
			Top 500	500
			Top 1000	1000
H.sapien	Doench et al. 2014	1278 [1022 train]	Full Matrix	6172	Raw + Onehot + QCT based on the H.sapien sgRNA sequence set from Doench et al. 2014	H. sapien	0.389120714	0.6525512
H.sapien	Chuai et al. 2018	16 749 [13 399 train]	Full Matrix	6172	Raw + Onehot + QCT based on the H.sapien sgRNA sequence set from Chuai et al. 2018	H. sapien	0.229489979	0.486193
H.sapien	Doench et al. 2014; Chuai et al. 2018	17 421 [13 936 train]	Full Matrix	6172	Raw + Onehot + QCT based on the H.sapien sgRNA sequence set from Doench et al. 2014 and Chuai et al. 2018	H. sapien	0.211671332	0.4964907
E.coli + H.sapien	Guo et al. 2016; Doench et al. 2014; Chuai et al. 2018	30000 [24000 train]	Full Matrix	6172	Raw + Onehot + QCT based on the E.coli sgRNA sequence set from Guo et al. 2016 and the H.sapien sgRNA sequence set sfrom Doench et al. 2014 and Chuai et al. 2018	E. coli + H. sapien	0.486194	0.6972761 [E.coli 0.504] [H.sapien 0.491]