Available via license: CC BY-ND 4.0
Content may be subject to copyright.
ClairS: a deep-learning method for long-
!
"
read somatic small variant calling
#
"
"
$
"
%&'()*+("%&'(,#-"./(&+0"1/#-"2'*"3&'(#-"4+(52+6"2''-"7+859+&"2+6-":/*;+(,"2/0<"
=
"
"
>
"
?'@+AB6'(B"0C"306@/B'A"1D*'(D'-"7&'"E(*F'AG*BH"0C"I0(,"J0(,-"I0(,"J0(,-"3&*(+"
K
"
"
L
"
#"7&'G'"+/B&0AG"D0(BA*;/B'M"'N/+OOH"B0"B&*G"P0A8"
Q
"
<"70"P&06"D0AA'G@0(M'(D'"G&0/OM";'"+MMA'GG'MR"S6+*OT"A;O/0UDGR&8/R&8"
V
"
"
!W
"
"
!!
"
Abstract
!#
"
XM'(YCH*(,"G06+YD"F+A*+(BG"*("B/60A"G+6@O'G"*G"+"DA/D*+O"B+G8-"P&*D&"*G"0Z'("@'AC0A6'M"/G*(,"
!$
"
GB+YGYD+O"6'B&0MG"+(M"&'/A*GYD"[OB'AG"+@@O*'M"B0"G&0AB5A'+M"M+B+R"I0P'F'A-"P*B&"B&'"*(DA'+G*(,"
!=
"
M'6+(M"C0A"O0(,5A'+M"G06+YD"F+A*+(B"D+OO*(,-"')*GY(,"6'B&0MG"&+F'"C+OO'("G&0ABR"70"+MMA'GG"B&*G"
!>
"
,+@-"P'"@A'G'(B"3O+*A1-"B&'"[AGB"M''@5O'+A(*(,5;+G'M-"O0(,5A'+M"G06+YD"G6+OO"F+A*+(B"D+OO'AR"
!K
"
3O+*A1"P+G"BA+*('M"0("6+GG*F'"GH(B&'YD"G06+YD"F+A*+(BG"P*B&"M*F'AG'"D0F'A+,'G"+(M"F+A*+(B"
!L
"
+OO'O'"CA'N/'(D*'G"\]^_`-"'(+;O*(,"*B"B0"+DD/A+B'OH"M'B'DB"+"P*M'"A+(,'"0C"G06+YD"F+A*+(BG"CA06"
!Q
"
@+*A'M"B/60A"+(M"(0A6+O"G+6@O'GR"9'"'F+O/+B'M"3O+*A1"/G*(,"B&'"O+B'GB"a+(0@0A'"b#Wc"
!V
"
I33!$V>5I33!$V>d2"M+B+G'BR"9*B&">W5C0OMe#>5C0OM"B/60Ae(0A6+O-"3O+*A1"+D&*'F'M"+"
#W
"
V$RW!feQKRQKf"@A'D*G*0(eA'D+OO"A+B'"C0A"1*(,O'"a/DO'0YM'"]+A*+Y0("\1a]G`-"+(M"KKR>=feKKRQVf"
#!
"
C0A"G06+YD"*(G'AY0(G"+(M"M'O'Y0(G"\X(M'OG`R"^@@OH*(,"3O+*A1"B0"G&0AB5A'+M"M+B+G'BG"CA06"
##
"
6/OY@O'"G0/AD'G"G&0P'M"D06@+A+;O'"0A";'g'A"@'AC0A6+(D'"B&+("1BA'O8+#"+(M"h/B'DB#R"i/A"
#$
"
[(M*(,G"G/,,'GB"B&+B"*6@A0F'M"A'+M"@&+G*(,"'(+;O'M";H"O0(,5A'+M"G'N/'(D*(,"*G"8'H"B0"+DD/A+B'"
#=
"
O0(,5A'+M"1a]"D+OO*(,-"'G@'D*+OOH"C0A"F+A*+(BG"P*B&"O0P"]^_R"7&A0/,&"')@'A*6'(BG"+DA0GG"F+A*0/G"
#>
"
D0F'A+,'-"@/A*BH-"+(M"D0(B+6*(+Y0("G'j(,G-"P'"M'60(GBA+B'M"B&+B"3O+*A1"*G"+"A'O*+;O'"G06+YD"
#K
"
F+A*+(B"D+OO'AR"3O+*A1"*G"0@'(5G0/AD'"+B"&g@GTee,*B&/;RD06eIJE5d^2e3O+*A1R"
#L
"
"
#Q
"
"
#V
"
Introduc.on
$W
"
^(+OHG*G"0C"D+(D'A",'(06'G"B&+B"*M'(YCH"+(M"D&+A+DB'A*k'"G06+YD"F+A*+(BG"&+G"'(+;O'M"+";'g'A"
$!
"
/(M'AGB+(M*(,"0C"B/60A"@A0,A'GG*0(1"+(M"O'M"B0"@A'D*G*0("0(D0O0,H2R"XM'(YCH*(,"G06+YD"
$#
"
F+A*+(BG-"&0P'F'A-"A'6+*(G"D&+OO'(,*(,"M/'"B0"*(BA+5"+(M"*(B'A5B/60A"&'B'A0,'('*BH-"P&*D&"0Z'("
$$
"
O'+MG"B0"O0P"]^_-"+(M"D0(C0/(M*(,"C+DB0AG-"*(DO/M*(,"G'N/'(D*(,"+AYC+DBG-"*(+M'N/+B'"
$=
"
G'N/'(D*(,"D0F'A+,'-"+(M"(0A6+O"D0(B+6*(+Y0(3R"S(M'+F0AG"P'A'"6+M'"B0"+MMA'GG"B&'G'"
$>
"
D&+OO'(,'G"+(M"6+)*6*k'"G'(G*YF*BH"+(M"+DD/A+DH"*("*M'(YCH*(,"G06+YD"F+A*+(BG"/G*(,"(')B5
$K
"
,'('A+Y0("G'N/'(D*(,"\al1`"G&0AB5A'+MG4-13R"I0P'F'A-"D0(GBA+*('M";H"A'+M"O'(,B&-"G&0AB"A'+MG"
$L
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
&+F'"O*6*B'M"F+A*+(B"M*GD0F'AH"D+@+;*O*BH"*("&+AM5B056+@",'(06*D"A',*0(G-"G/D&"+G"
$Q
"
&060@0OH6'AG"+(M"G',6'(B+O"M/@O*D+Y0(GR"7&*G"@A0;O'6"*G"')@'DB'M"B0";'"+OO'F*+B'M"B&A0/,&"
$V
"
O0(,5A'+M"G'N/'(D*(,14R"i)C0AM"a+(0@0A'"7'D&(0O0,*'G"\ia7`"*G"+60(,"B&'"O'+M*(,"O0(,5A'+M"
=W
"
G'N/'(D*(,"B'D&(0O0,*'G"+(M"0m'AG"6*(*+B/A*k'M"G'N/'(D*(,"M'F*D'G"+(M"C+GB"G+6@O'5B05M+B+"
=!
"
B/A(+A0/(M-"P&*D&"*G"+"DA'M*;O'"GB'@"B0P+AMG"M'60DA+Yk*(,"G'N/'(D*(,";H"MA+GYD+OOH"A'M/D*(,"
=#
"
B&'"D0GB"0C"D+AAH*(,"0/B"G'N/'(D*(,"')@'A*6'(BGR"ia7"A+P"A'+MG"P'A'"A'@0AB'M"B0"&+F'"+("'AA0A"
=$
"
A+B'"0C"$5!>f"*("B&'"@+GB15R"7&*G"P+G"A'M/D'M"B0"!f"0A"O0P'A"/G*(,"ia7nG"O+B'GB"b#Wc"
==
"
D&'6*GBAH16R"7&'",+@"*G"GYOO"G*,(*[D+(B-"&0P'F'A-"D06@+A'M"P*B&"al1"G&0AB5A'+MG-"P&*D&"&+F'"+("
=>
"
+F'A+,'"'AA0A"A+B'"+B"WR!f17-"6+8*(,"B&'"G06+YD"F+A*+(B"D+OO'AG"0(D'"M'G*,('M"C0A"G&0AB"A'+MG"
=K
"
@A+DYD+OOH"/(P0A8+;O'"C0A"ia7"O0(,5A'+MGR"
=L
"
"
=Q
"
l'A6O*('"F+A*+(BG"+A'"0Z'("D0(G*M'A'M"B0";'"'+G*'A"B0"D0AA'DBOH"*M'(YCH"B&+("G06+YD"F+A*+(B"
=V
"
D+OO*(,R"7&'"[AGB"+g'6@B"B0"D+OO",'A6O*('"G6+OO"F+A*+(BG"/G*(,"(0*GH"ia7"O0(,5A'+MG"P+G"6+M'";H"
>W
"
3O+*AF0H+(B'"*("#W!Q18R"7&'"P0A8"P+G"'(+;O'M"/G*(,"!`"+"M''@"('/A+O"('BP0A8-"P&*D&"P+G"[AGB"
>!
"
/G'M"C0A"F+A*+(B"D+OO*(,";H"?''@]+A*+(B19o"+(M"#`"B&'"&*,&5N/+O*BH"8(0P("BA/B&"F+A*+(BG"*("lX^d"
>#
"
A'C'A'(D'"G+6@O'G"C0A"('/A+O"('BP0A8"60M'O"BA+*(*(,20R"1/;G'N/'(B"P0A8G";H"3O+*AF0H+(B'"
>$
"
*(DO/M*(,"3O+*A21"+(M"3O+*A$22-"*(BA0M/D'M"0@Y6*k'M"('BP0A8"*(@/B-"('BP0A8"0/B@/B-"('BP0A8"
>=
"
+AD&*B'DB/A'-"+(M"P0A8p0P"M'G*,(G"B0"6+8'"B&'";'GB"0/B"0C"(0*GH"ia7"M+B+"C0A",'A6O*('"G6+OO"
>>
"
F+A*+(B"D+OO*(,R"d0B&"3O+*A$"+(M"+"@*@'O*('"(+6'M"qSqqS:5h+A,*(5?''@]+A*+(B23"\?''@]+A*+(B`-"
>K
"
P&*D&"*G"+OG0"M'G*,('M"C0A"ia7"O0(,5A'+M",'A6O*('"G6+OO5F+A*+(B"D+OO*(,-"&+F'"M'60(GBA+B'M"
>L
"
;'g'A"G*(,O'"(/DO'0YM'"@0OH60A@&*G6"\1aq`5D+OO*(,"@'AC0A6+(D'"B&+("/G*(,"B&'"G+6'"D0F'A+,'"
>Q
"
0C"XOO/6*(+"G&0AB"A'+MGR"I0P'F'A-"P&*O'"G0O/Y0(G"+A'"A'+MH"C0A"ia7"O0(,5A'+M",'A6O*('"G6+OO5
>V
"
F+A*+(B"D+OO*(,-"B&'A'"&+G";''("(0"D+OO'A"+F+*O+;O'"C0A"ia7"O0(,5A'+M"G06+YD"G6+OO5F+A*+(B"
KW
"
D+OO*(,R"9'"(0B'"B&+B"ia7"O0(,5A'+M"G06+YD"1]"\GBA/DB/A+O"F+A*+(B`"D+OO'AG-"*(DO/M*(,"1(*r'G#24"
K!
"
+(M"a+(060(GF25-"P&*D&"P'A'"M'F'O0@'M"*("B&'"@+GB"H'+A"D+OO'M"C0A"B&'"M'F'O0@6'(B"0C"+"G6+OO"
K#
"
F+A*+(B"D+OO'A"B0"D06@O'B'"B&'"ia7"O0(,5A'+M"G06+YD"F+A*+(B5D+OO*(,"P0A8p0PR"
K$
"
"
K=
"
E(C0AB/(+B'OH-"G06'"M'G*,(G"DA*YD+O"B0"ia7"O0(,5A'+M",'A6O*('"F+A*+(B"D+OO*(,"+A'"(0B"+@@O*D+;O'"
K>
"
B0"G06+YD"F+A*+(B"D+OO*(,R"_*AGB-"*("B&'*A"('BP0A8"0/B@/B-";0B&"3O+*A$"+(M"?''@]+A*+(B"+@@OH"+"
KK
"
GBA0(,"M*@O0*M",'(06'"+GG/6@Y0(R"3O+*A$"/G'G"+"#!5,'(0BH@'"0/B@/B-"P&*D&"*G"+"BP05
KL
"
D06;*(+Y0("0C"^-"3-"l-"7-"*(G'AY0(-"+(M"M'O'Y0(22R"?''@]+A*+(B"/G'G"+"B&A''5D+B',0AH"0/B@/B"
KQ
"
B&+B"*(DO/M'G"&065A'C"\&060kH,0/G"A'C'A'(D'`-"&'B"\&'B'A0kH,0/G`-"+(M"&065+OB"\&060kH,0/G"
KV
"
+OB'A(+YF'`23R"d0B&"3O+*A$"+(M"?''@]+A*+(B"+A'"DO+GG*[D+Y0("60M'OG"B&+B"/G'"0;G'AF'M"+OO'O'"
LW
"
CA'N/'(DH"0C"+OB'A(+YF'"+OO'O'G"+G"('BP0A8"*(@/B-"+(M"0/B@/B"B&'"D+B',0AH"B&+B"A'@A'G'(BG"B&'"
L!
"
')@'DB'M"+OO'O'"CA'N/'(DH"0C"+"F+A*+(B"\'R,R-"+OO'O'"CA'N/'(DH"W-"WR>-"!-"+(M"WR>eWR>"C0A",'(0BH@'"
L#
"
WeW-"We!-"!e!-"+(M"!e#-"A'G@'DYF'OH`R"I0P'F'A-"G06+YD"F+A*+(BG"&+F'"]^_"A+(,*(,"D0(Y(/0/GOH"
L$
"
CA06"W"B0"!R"9*B&0/B"+"D'AB+*("@O0*MH-"G06+YD"F+A*+(B"D+(M*M+B'G"&+F'"(0"')@'DB'M"+OO'O'"
L=
"
CA'N/'(DH"C0A"+"60M'O"B0"B'GB"+,+*(GBR"7&/G-"+"('P"M'G*,("*G"A'N/*A'MR"^G"+("')+6@O'-"+"('P"
L>
"
M'G*,("D0/OM"/G'"+"A',A'GG*0("60M'O"B0"M'A*F'"]^_"M*A'DBOH-"0A"+"DO+GG*[D+Y0("60M'O"B0"G*6@OH"
LK
"
M'B'A6*('"P&'B&'A"+"D+(M*M+B'"*G"+"G06+YD"F+A*+(B"0A"(0B"+(M"*(C'A"]^_"G/;G'N/'(BOHR"1'D0(M-"
LL
"
B&'"G'F'("GB+(M+AM"lX^d"A'C'A'(D'"G+6@O'G"IlWW!sIlWWL"@A0F*M'"+@@A0)*6+B'OH"#>"6*OO*0("
LQ
"
BA/B&",'A6O*('"F+A*+(BG20-"P&*D&"+A'"DA*YD+O"C0A"B&'"60M'O"BA+*(*(,"0C"+(H"GB+B'50C5B&'5+AB-"M''@5
LV
"
O'+A(*(,5;+G'M",'A6O*('"F+A*+(B"D+OO'AGR"I0P'F'A-"*("B'A6G"0C"8(0P("BA/B&"G06+YD"F+A*+(BG-"0(OH"
QW
"
B&'"I33!$V>sI33!$V>d2"\+"&/6+("BA*@O'5(',+YF'";A'+GB"D+(D'A"D'OO"O*('"+(M"+"(0A6+O"D'OO"O*('"
Q!
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
M'A*F'M"CA06"B&'"d"OH6@&0DHB'G"0C"B&'"G+6'"M0(0A-"&'A'+Z'A"A'C'AA'M"B0"+G"I33!$V>ed2`"
Q#
"
B/60A5(0A6+O"@+*A"P+G"@/;O*G&'M";H"B&'"106+YD"h/B+Y0("90A8*(,"lA0/@"0C"B&'"1Sb3#"
Q$
"
\1'N/'(D*(,"b/+O*BH"30(BA0O"q&+G'"XX`"D0(G0AY/63R"XB"D0(B+*(G"0(OH"$V->KW"1a]G"+(M"!-V##"
Q=
"
X(M'OG-"P&*D&"*G"0AM'AG"0C"6+,(*B/M'"C'P'A"B&+("B&'"+F+*O+;O'"BA/B&",'A6O*('"F+A*+(BG-"+(M"C+A"
Q>
"
CA06"'(0/,&"C0A"M''@"('/A+O5('BP0A8"60M'O"BA+*(*(,R"X("B&'"+;G'(D'"0C"+M'N/+B'"A'+O"B/60A5
QK
"
(0A6+O"G+6@O'G-"0('"D0/OM"B&*(8"0C"GH(B&'YD"M+B+"+G"+"G0O/Y0(R"d+6G/A,'0("G@*8'M"G06+YD"
QL
"
F+A*+(BG"*(B0"G'N/'(D*(,"A'+MG"B0"6*6*D"+"B/60A26R"7&*G"P+G"+"G/DD'GGC/O"6'B&0M"C0A"G&0AB"
QQ
"
A'+MG-";/B"*B"M0'G(nB"P0A8"C0A"O0(,"A'+MG"C0A"BP0"A'+G0(GT"!`"XOO/6*(+"G&0AB"A'+MG"+A'"A'+M"0('"
QV
"
;+G'"+B"+"Y6'-"P&'A'+G"ia7"O0(,"A'+MG"+A'"A'+M"+G"B&'"G*,(+OG"0C"+"GO*M*(,">56'A"0A"K56'A"
VW
"
P*(M0PR"7&+B"*G-"+"G@*8'5*("F+A*+(B"+OG0"D&+(,'G"B&'"G*,(+OG"0C"B&'"+Mt+D'(B";+G'GR"d+G'G"D+(";'"
V!
"
;+G'5D+OO'M"CA06"G*,(+OG";/B"D+((0B";'"+/B&'(YD+OOH"B/A('M";+D8"*(B0"G*,(+OG-"G0"B&'"G@*8'5*("
V#
"
6'B&0M"D+((0B";'"+@@O*'M"B0"O0(,"A'+MGo"+(M"#`"d+6G/A,'0("M0'G"(0B"+@@OH"B&'"@0*(B"B&+B"+"
V$
"
G06+YD"F+A*+(B"*G"/G/+OOH"C0/(M"*("0(OH"0('"&+@O0BH@'"\'*B&'A"6+B'A(+O"0A"@+B'A(+O`";/B"*G"
V=
"
6*GG*(,"*("+(0B&'AR"7&'"/G'"0C"O0(,"A'+MG"0F'A"G&0AB"A'+MG"C0A"G06+YD"F+A*+(B"D+OO*(,"6+8'G"
V>
"
G'(G'"0(OH"*C"B&'"O0(,5A'+M"+MF+(B+,'"0C"&+@O0BH@*(,"\+OG0"D+OO'M"@&+G*(,`"*G"/YO*k'MR"^OO"B&*(,G"
VK
"
D0(G*M'A'M-"+"('P"6'B&0M"C0A"GH(B&'G*k*(,"O0(,5A'+M"M+B+"B0"D0(B+*("+;/(M+(B"@O+/G*;O'"
VL
"
G06+YD"F+A*+(BG"*G"(''M'MR"
VQ
"
"
VV
"
X("B&*G"GB/MH-"P'"@A'G'(B"3O+*A5106+YD"\3O+*A1`-"B&'"[AGB"G06+YD"G6+OO"F+A*+(B"D+OO'A"C0A"ia7"O0(,"
!WW
"
A'+MG-"P&*D&"*G"(+6'M"+Z'A"*BG",'A6O*('"F+A*+(B"D+OO'A"@A'M'D'GG0AGR"3O+*A1"MA+PG"0("B&'"
!W!
"
G/DD'GGC/O"')@'A*'(D'"0C"B&'"3O+*A"G'A*'G-"+(M"/G'G"+"('P"('BP0A8"0/B@/B"+(M"+"('P"P0A8p0P"
!W#
"
M'G*,("B0"+MMA'GG"B&'"D0(Y(/0/G"]^_"G@+D'"0C"G06+YD"F+A*+(BGR"dH"D0(G*M'A*(,"BP0"M*m'A'(B"
!W$
"
G+6@O'G-"^"+(M"d-"+G"B/60A"+(M"(0A6+O-"A'G@'DYF'OH-"+(M"M''6*(,"+",'A6O*('"F+A*+(B"G@'D*[D"B0"
!W=
"
^"+G"+"G06+YD"F+A*+(B"+,+*(GB"d-"P'"M'F*G'M"+"M+B+"GH(B&'YD"GBA+B',H"B&+B"/G'G"0(OH"B&'"A'+O"
!W>
"
O0(,"A'+MG"0C"lX^d"A'C'A'(D'"G+6@O'G"P*B&"8(0P(",'A6O*('"F+A*+(BG-";/B"D+("G*6/O+B'"G06+YD"
!WK
"
F+A*+(BG"0C"+(H"B/60A"@/A*BH-"G'N/'(D*(,"D0F'A+,'"\O0P'A"B&+("B&'"M+B+"G0/AD'`-"+(M"O'F'O"0C"
!WL
"
(0A6+O"D0(B+6*(+Y0(R"7&'"GBA+B',H"D+("B&'0A'YD+OOH"@A0M/D'"+("*([(*B'"(/6;'A"0C"G06+YD"
!WQ
"
F+A*+(BG"C0A"60M'O"BA+*(*(,R"9'"G&0P"A'G/OBG"B0"&*,&O*,&B"&0P"@&+G*(,"*6@A0F'G"G06+YD"F+A*+(B5
!WV
"
D+OO*(,"@'AC0A6+(D'"0("O0(,"A'+MGR"70"O'F'A+,'"60A'"A'60B'"+O*,(6'(B"*(C0A6+Y0("B&+B"*G"
!!W
"
D06@/B+Y0(+OOH"*6@A+DYD+O"B0"*(DO/M'"*("B&'"('BP0A8"*(@/B-"P'"M'F*G'M"+"@0GB5@A0D'GG*(,"GB'@"
!!!
"
B&+B"G'+AD&'G"C0A"+(D'GBA+O"&+@O0BH@'"G/@@0AB"C0A"+(H"G06+YD"F+A*+(B"D+(M*M+B'GR"7&*G"GB'@"
!!#
"
A'60F'M"+"D0(G*M'A+;O'"+60/(B"0C"C+OG'"@0G*YF'"D+OOG"*("0/A"')@'A*6'(BGR"_0A";'(D&6+A8*(,-"
!!$
"
P'"G'N/'(D'M"*("B0B+O"L>5C0OM"I33!$V>"+(M"=>5C0OM"I33!$V>d2"ia7"b#Wc"O0(,5A'+MG"\M+B+"
!!=
"
M'@0G*B'M"B0"a3dX"1:^`-"+(M"/G'M"B&'"BA/B&"G06+YD"F+A*+(BG"@A0F*M'M";H"B&'"1Sb3#"
!!>
"
D0(G0AY/63R"9*B&">W5C0OMe#>5C0OM"B/60Ae(0A6+O-"3O+*A1"+D&*'F'M"QKRQKfeV$RW!f"
!!K
"
A'D+OOe@A'D*G*0("A+B'"1a]G-"+(M"KKRQVfeKKR>=f"C0A"G06+YD"X(M'OG"P&'("B+A,'Y(,"]^_"uWRW>R"_0A"
!!L
"
F+A*+(BG"P*B&"]^_"uWR#-"B&'"(/6;'AG",0"/@"B0"V=RK>feVKRK$f"C0A"1a]G-"+(M"L$R##feLLR$>f"C0A"
!!Q
"
G06+YD"X(M'OGR"9'"+OG0"G&0P"B&'"@'AC0A6+(D'"0C"3O+*A1"+B"M*m'A'(B"B/60Ae(0A6+O"D0F'A+,'G-"
!!V
"
B/60A"@/A*BH"+(M"(0A6+O"D0(B+6*(+Y0(R"3O+*A1"*G"M'G*,('M"C0A"ia7"O0(,"A'+MG-";/B"B&'"P&0O'"
!#W
"
6'B&0M"*G"+OG0"+@@O*D+;O'"B0"XOO/6*(+"G&0AB"A'+MGR"7&*G"F'AG+YO*BH"+OO0P'M"/G"B0";'(D&6+A8"3O+*A1"
!#!
"
+,+*(GB"GB+B'50C5B&'5+AB"G&0AB5A'+M"G06+YD"F+A*+(B"D+OO'AG"D0(G*M'A*(,"B&+B"B&'A'"*G"(0"0B&'A"
!##
"
O0(,5A'+M"G06+YD"F+A*+(B"D+OO'A"B0";'(D&6+A8"+,+*(GBR"7&'"A'G/OBG"G&0P"B&+B"3O+*A1"@'AC0A6'M"
!#$
"
D06@+A+;OH"0A"GO*,&BOH";'g'A"B&+("B&'"D/AA'(B"&'/A*GYD5;+G'M"+(M"M''@5O'+A(*(,5;+G'M"D+OO'AG"
!#=
"
0("G&0AB"A'+MGR"
!#>
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
"
!#K
"
"
!#L
"
Results
!#Q
"
The ClairS method
!#V
"
7&'"3O+*A1"6'B&0M"*G"'O+;0A+B'M"*("B&'"h'B&0M"G'DY0(R"7')BG"*(";0OM"*("B&*G"+(M"B&'"C0OO0P*(,"
!$W
"
BP0"@+A+,A+@&G"D+(";'"C0/(M"+G"G/;G'DY0("YBO'G"*("B&'"h'B&0M"G'DY0(R"X("B&'"/G'"0C"M''@5
!$!
"
O'+A(*(,"C0A"O0(,5A'+M"G06+YD"F+A*+(B"D+OO*(,-"3O+*A1"&+G"6+M'";A'+8B&A0/,&G"*("!"#$%$%&'(#)#'
!$#
"
*+%),-*$*"+(M"B&'"./#$"0"12"3421'#%('(-*$&%R"
!$$
"
"
!$=
"
:',+AM*(,"BA+*(*(,"M+B+"GH(B&'G*G-"_*,/A'"!+"G&0PG"B&'"P0A8p0P"C0A"5-%-"#6%&'*+%),-67')892"'
!$>
"
#%('*+%),-67'%2"9#/R"^G"P'"A'N/*A'M"G+6@O*(,"P*B&0/B"A'@O+D'6'(B-"+(M"*("F*'P"0C"B&'"
!$K
"
D0660("@A+DYD'"B&+B"+"B/60A"G+6@O'"&+F*(,"&*,&'A"D0F'A+,'"B&+("*BG"6+BD&*(,"(0A6+O-"P'"
!$L
"
,+F'".2:-"#&-'#(:$7-';2"'),-'*28"7-'(#)#R"?'B+*OG"0C"&0P"&060kH,0/G"+(M"&'B'A0kH,0/G"
!$Q
"
,'A6O*('"F+A*+(BG"*("M*m'A'(B"G+6@O'G"+A'"D0(F'AB'M"*(B0"G06+YD"F+A*+(BG"+A'",*F'("*("<-"$:$%&'
!$V
"
98/6=/-'7#)-&2"$-*'2;':#"$#%)*';"29'#'*+%),-67')892">%2"9#/'=#$"R"1'B/@G"B0"+F0*M"@A+DYD+O"
!=W
"
D0(D'A(G"+(M"O*6*B+Y0(G"0C"B&'"M+B+"GH(B&'G*G"6'B&0M"+A'",*F'("*("?),-"'(-)#$/*'#@28)'),-'
!=!
"
:#"$#%)*'*-/-7)-(';2"'92(-/')"#$%$%&R"i/A"M+B+"GH(B&'G*G"6'B&0M"*G";+G'M"0("B&'"0;G'AF+Y0("
!=#
"
B&+B"+("+/B&'(YD"G06+YD"F+A*+(B"*G"/G/+OOH"C0/(M"*("B&'"A'+MG"0C"+"G*(,O'"&+@O0BH@'"\M'@*DB'M"*("
!=$
"
S)B'(M'M"?+B+"_*,/A'"!+`R"^OG0-"P'"C0/(M"B&+B"A,#*$%&'$%;2"9#62%'-%,#%7-*'*29#67':#"$#%)'
!==
"
7#//$%&'=-";2"9#%7-"*("3O+*A1R"
!=>
"
"
!=K
"
:',+AM*(,"B&'"3O+*A1"P0A8p0P"+(M"M'G*,(-"_*,/A'"#+"G&0PG"+("?:-":$-1"0C"B&'"3O+*A1"P0A8p0PR"
!=L
"
_*,/A'"#;"G&0PG"0)-='BC'5-"9/$%-':#"$#%)'7#//$%&D'=,#*$%&'#%('"-#(',#=/2)#&&$%&-"+(M"_*,/A'"#D"
!=Q
"
G&0PG"0)-='EF'A$/-8=G@#*-('#%(';8//G#/$&%9-%)'@#*-(':#"$#%)'7#//$%&R"_0A"'+D&"G+6@O'-"P&*O'"
!=V
"
0(OH"+"CA+DY0("0C",'(06'"@0G*Y0(G"&+G"+OB'A(+YF'"+OO'O'"G/@@0AB"+(M"+60(,"B&'6-"0(OH"+"C'P"
!>W
"
&+F'"B&'"@0B'(Y+O"B0";'"D+OO'M"G06+YD"F+A*+(BG-"P'"/G'M"&'/A*GYDG"C0A"0-/-76%&':#"$#%)'
!>!
"
7#%($(#)-*R"?'B+*OG"0C"!,-'(-*$&%'2;'=$/-8='$%=8)'#%(';8//G#/$&%9-%)'$%=8)"+(M"!,-'(-*$&%'2;'
!>#
"
%-8"#/'%-)12"3*"+A'"G&0P("*("_*,/A'"$R"7&'"P+HG"*("P&*D&"3O+*A1"M*m'AG"CA06"*BG"@A'M'D'GG0A-"
!>$
"
3O+*A$-"+A'"M*GD/GG'M"*("h'B&0MR"_*,/A'"#M"G&0PG"0)-='HF'0-#"7,';2"'#%7-*)"#/',#=/2)+=-'
!>=
"
*8==2")-"P&*D&"*G"+"@0GB5@A0D'GG*(,"GB'@"B&+B"O'F'A+,'G"60A'"A'60B'"+O*,(6'(B"*(C0A6+Y0("B0"
!>>
"
G'+AD&"C0A"+(D'GBA+O"&+@O0BH@'"G/@@0AB"B0"B&'"G06+YD"F+A*+(BG"D+OO'M"*("GB'@"#R"70"+M+@B"B0"
!>K
"
M*m'A'(B"/G+,'"GD'(+A*0G-"6/OY@O'"?8)=8)"0@Y0(G"+A'"@A0F*M'MR"
!>L
"
"
!>Q
"
ClairS performance on ONT data
!>V
"
^"G/66+AH"0C"B&'"ia7"M+B+"/G'M"C0A"60M'O"BA+*(*(,"+(M";'(D&6+A8*(,"*G"G&0P("*("
!KW
"
1/@@O'6'(B+AH"7+;O'"!R"9'"BA+*('M"B&'"3O+*A1"ia7"60M'O"/G*(,"GH(B&'YD"M+B+",'('A+B'M"CA06"
!K!
"
BP0"lX^d"G+6@O'GT"IlWW!"+(M"IlWW#20R"9'"/G'M";0B&"IlWW!eIlWW#"+(M"IlWW#eIlWW!"+G"
!K#
"
B/60Ae(0A6+O"G+6@O'G"C0A"M+B+"GH(B&'G*GR"7&'"IlWW#"G+6@O'"&+G"LKR#V5C0OM"D0F'A+,'"+(M"P+G"
!K$
"
6+M'"+F+*O+;O'";H"a+(0@0A'"B&A0/,&"SqX#hS"2+;GR"7&'"IlWW!"G+6@O'"&+G"=QR==5C0OM"D0F'A+,'"
!K=
"
+(M"P+G"G'N/'(D'M"+B"IJE-"P*B&"M'B+*OG",*F'("*("B&'"h'B&0M"s"ia7"O*;A+AH"@A'@+A+Y0("+(M"
!K>
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
G'N/'(D*(,"G'DY0(R"_0OO0P*(,"B&'"D0(F'(Y0("0C"B&'"GB+B'50C5B&'5+AB"M''@5O'+A(*(,5;+G'M"
!KK
"
F+A*+(B"D+OO'AG-"P'"')DO/M'M"A'+MG"+(M"F+A*+(BG"CA06"3&A060G06'"#W"CA06"60M'O"BA+*(*(,R""
!KL
"
"
!KQ
"
_0A";'(D&6+A8*(,-"P'"/G'M"B&'"I33!$V>ed2"B/60A5(0A6+O"@+*A-"P*B&"8(0P("BA/B&"F+A*+(BG"
!KV
"
@A0F*M'M";H"B&'"1Sb3#"D0(G0AY/63R"7&'"I33!$V>"G+6@O'"&+G"L>RVL5C0OM"D0F'A+,'R"7&'"
!LW
"
I33!$V>d2"G+6@O'"&+G"=>R>>5C0OM"D0F'A+,'R"d0B&"G+6@O'G"P'A'"G'N/'(D'M"+B"BP0"G'N/'(D*(,"
!L!
"
D'(B'AG"\IJE"+(M"a0F0,'('`"C0A"N/+O*BH"D0(BA0O"@/A@0G'GR"7&'"H*'OM"0C"B&'"BP0"D'(B'AG"*G"G&0P("
!L#
"
*("1/@@O'6'(B+AH"7+;O'"#R"i(OH"1Sb3#"BA/B&"F+A*+(BG"O+;'O'M"vI*,&30(Cw"\&*,&"D0([M'(D'`"+(M"
!L$
"
vh'M30(Cw"\6'M*/6"D0([M'(D'`"P'A'"/G'M"C0A";'(D&6+A8*(,R"9&*O'"B&'A'"+A'"x=W8"1a]G";/B"
!L=
"
0(OH"x#8"X(M'OG"*("B&'"BA/B&"G'B-"B&'";*,"M*m'A'(D'"*("B&'"(/6;'A"0C"BA/B&"F+A*+(BG"A'G/OBG"*("
!L>
"
M*m'A'(B"+(+OHYD+O"@0P'A";'BP''("B&'"BP0"F+A*+(B"BH@'GR"10"P'"[AGB";'(D&6+A8'M"6/OY@O'"
!LK
"
D0F'A+,'G-"B/60A"@/A*BH"+(M"(0A6+O"D0(B+6*(+Y0("P*B&"0(OH"1a]G-"+(M"B&'(";'(D&6+A8'M"+(M"
!LL
"
M*GD/GG"G06+YD"X(M'O5D+OO*(,"@'AC0A6+(D'"*("+"G'@+A+B'"G'DY0(R"9'"/G'M"XOO/6*(+nG"I+@O0BH@'"
!LQ
"
306@+A*G0("700OG27"B0",'('A+B'"B&'"@'AC0A6+(D'"[,/A'G-"*(DO/M*(,"_!51D0A'-"qA'D*G*0(-"+(M"
!LV
"
:'D+OO-"+(M"DA0GG5F+O*M+B'M"B&'6"P*B&"B&'"vD06@+A'yFDCw"G/;60M/O'"*("3O+*A1R"h0A'"
!QW
"
DO+A*[D+Y0(G"+(M"@+A+6'B'AG"+A'"G&0P("*("B&'"h'B&0M"s"d'(D&6+A8*(,"G'DY0(R""
!Q!
"
"
!Q#
"
_0A";0B&"60M'O"BA+*(*(,"+(M";'(D&6+A8*(,-"P'"/G'M"l:3&$Q-"P&*D&"*G"B&'"('P'GB"A'C'A'(D'"
!Q$
"
,'(06'"F'AG*0("0("P&*D&";0B&"lX^d"+(M"1Sb3#"BA/B&"F+A*+(BG"+A'";+G'MR"^OO"ia7"G'N/'(D*(,"
!Q=
"
M+B+"P'A'";+G'5D+OO'M"/G*(,"l/@@H"F'AG*0("KR!R>"+(M"+O*,('M"B0"l:3&$Q"/G*(,"6*(*6+@#"F'AG*0("
!Q>
"
#R!L5AV=!R"7&'"D066+(M"O*('"/G'M"*G",*F'("*("B&'"1/@@O'6'(B+AH"a0B'G"s"3066+(M"O*('G"/G'M"
!QK
"
G'DY0(R"^OO"M+B+"6'(Y0('M"+;0F'-"*(DO/M*(,"ia7"G'N/'(D*(,"M+B+-"lX^d"BA/B&"F+A*+(BG-"1Sb3#"
!QL
"
BA/B&"F+A*+(BG-"+(M"A'C'A'(D'",'(06'G-"+A'"@/;O*DOH"+DD'GG*;O'"F*+"O*(8G"0A"1:^"+DD'GG*0("X?G"
!QQ
"
O*GB'M"*("B&'"1/@@O'6'(B+AH"a0B'G"s"?+B+"+F+*O+;*O*BH"G'DY0(R"
!QV
"
"
!VW
"
!"#$%#&'()"*+,-.*/,0"#"(-*-1&%#*)%&2,('3%(4*'(/*(%#&'5*)%6"#'7"8"9'"+GG'GG'M"B&'"3O+*A1"
!V!
"
@'AC0A6+(D'"P*B&"M*m'A'(B"D06;*(+Y0(G"0C"B/60AG"+(M"(0A6+O"D0F'A+,'R"9'"B'GB'M"B&A''"
!V#
"
B/60A"D0F'A+,'"A+B'GT"#>5-">W5-"+(M"L>5C0OMR"9'"+@@O*'M"#>5C0OM"+G"B&'"[AGB"GB'@"+G"*B"A'@A'G'(BG"
!V$
"
+"D0(G'AF+YF'"B&A0/,&@/B"'GY6+Y0("0C"+(":!WR=R!"qA06'B&Xia"p0PD'OOR"9'"+OG0"B'GB'M"B&A''"
!V=
"
(0A6+O"D0F'A+,'GT"#W5-"#>5-"+(M"$W5C0OMR"7&'"#>5C0OM"GB'@"A'G'6;O'G"B&'"B&A0/,&@/B"F+A*+(D'"0C"
!V>
"
+"G*(,O'"p0PD'OOR"i/A"')@'A*6'(BG"+*6'M"B0"*6*B+B'"+"@A+DYD+O"G'j(,"C0A"DO*(*D+O"D+(D'A"
!VK
"
M*+,(0G*G-"P*B&"B&'"B/60A"G+6@O'"D0F'A+,'"*(DA'+G'M"0('"p0PD'OO"+B"+"Y6'"B0"G''8"&*,&'A"
!VL
"
M*GD0F'AH"@0P'A-";/B"B&'"(0A6+O"G+6@O'"D0F'A+,'"*G"[)'M"+B"+"G*(,O'"p0PD'OO"C0A"D0GB5
!VQ
"
'm'DYF'('GGR""
!VV
"
"
#WW
"
7&'"A'G/OBG"+A'"G&0P("+G"qA'D*G*0(5:'D+OO"D/AF'G"*("_*,/A'"=+R"7&'"'F+O/+Y0("6'BA*DG"+B"BP0"
#W!
"
F+A*+(B"N/+O*BH"D/B0mG"\Q"+(M"!>`"+A'"G&0P("*("1/@@O'6'(B+AH"7+;O'"$R"b/+O*BH"D/B0m"!>"
#W#
"
\&'A'+Z'A"A'C'AA'M"B0"+G"v@A*0A*Yk'5C!"60M'w`"[OB'AG"60A'"F+A*+(BG"+(M"+*6G"C0A";+O+(D'M"
#W$
"
@A'D*G*0("+(M"A'D+OOR"b/+O*BH"D/B0m"Q"\v@A*0A*Yk'5A'D+OO"60M'w`"A'B+*(G"60A'"F+A*+(BG"+(M"+*6G"
#W=
"
C0A"&*,&'A"A'D+OOR"X("B&'"@A*0A*Yk'5C!"60M'-"P*B&"(0A6+O"D0F'A+,'"[)'M"+B"#>5C0OM-"3O+*A1"+D&*'F'M"
#W>
"
V>RW$feLQRL!feQKR!!f-"V$RW!feQKRQKfeQVRQ$f-"+(M"V#RV=feQKRV#feQVRQ$f"
#WK
"
@A'D*G*0(eA'D+OOeC!5GD0A'"+B"#>5-">W5-"+(M"L>5C0OM"B/60A"D0F'A+,'-"A'G@'DYF'OHR"_A06"#>5"B0">W5
#WL
"
C0OM-"B&'"A'D+OO"*(DA'+G'M"CA06"LQR!!f"B0"QKRQKf"\cQRL>f`-"P*B&"+"#RW#f"@A'D*G*0("MA0@"\CA06"
#WQ
"
V>RW$f"B0"V$RW!f`R"_A06">W5"B0"L>5C0OM-"&0P'F'A-"(0"*6@A0F'6'(B"P+G"0;G'AF'MR"X("B&'"
#WV
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
@A*0A*Yk'5A'D+OO"60M'-"+OG0"P*B&"(0A6+O"D0F'A+,'"[)'M"+B"#>5C0OM-"3O+*A1"+D&*'F'M"
#!W
"
Q#RKKfeV!R>>feQKRQQf-"LWR#!feVKR!WfeQ!R!=f-"+(M"K$RQWfeVKRKWfeLKRQ>f"
#!!
"
@A'D*G*0(eA'D+OOeC!5GD0A'"+B"#>5-">W5-"+(M"L>5C0OM"B/60A"D0F'A+,'-"A'G@'DYF'OHR"306@+A'M"B0"
#!#
"
@A*0A*Yk'5C!-"B&'"@A*0A*Yk'5A'D+OO"60M'"A'+D&'M"V!R>>f"A'D+OO"+B"#>5C0OM"\+,+*(GB"LQRL!f-"
#!$
"
c!#RQ=f`-"+(M"VKR!Wf"+B">W5C0OM"\+,+*(GB"QKRQKf-"cVR#=f`R"^OB&0/,&"3O+*A$"*G"(0B"M'G*,('M"C0A"
#!=
"
G06+YD"F+A*+(BG-"B0",*F'"+"A'C'A'(D'"@0*(B-"P'"D0(M/DB'M"')@'A*6'(BG"P*B&"*B"/G*(,"B&'"G+6'"
#!>
"
M+B+G'BG";H"D0(G*M'A*(,"B&'",'A6O*('"F+A*+(BG"C0/(M"*("B/60A";/B"(0B"*("(0A6+O"+G"G06+YD"
#!K
"
F+A*+(BGR"^B"#>5C0OM"(0A6+O-"3O+*A$"&+M"+"!VR#LfeL#R!Wfe$WR=#f-"#VRQ#feKQR$#fe=!R>#f-"+(M"
#!L
"
$KR#!feK=R>Kfe=KR=Wf"@A'D*G*0(eA'D+OOeC!5GD0A'"+B"#>5-">W5-"+(M"L>5C0OM"B/60A"D0F'A+,'R"7&'"
#!Q
"
A'G/OBG"&*,&O*,&B"B&'"*(+@@A0@A*+B'('GG"0C"/G*(,"+",'A6O*('"F+A*+(B"D+OO'A"C0A"G06+YD"F+A*+(B"
#!V
"
D+OO*(,R"
##W
"
"
##!
"
X(";0B&"@A*0A*Yk'5C!"+(M"@A*0A*Yk'5A'D+OO"60M'G-"A+*G*(,"B&'"(0A6+O"D0F'A+,'"D0(G*GB'(BOH"
###
"
*(DA'+G'M"G06+YD"F+A*+(B5D+OO*(,"@'AC0A6+(D'R"9*B&"B/60A"D0F'A+,'"[)'M"+B">W5C0OM"+(M"/G*(,"
##$
"
B&'"@A*0A*Yk'5C!"60M'-"3O+*A1"+D&*'F'M"+"V!RQKfeQ$RLLfeQLRK$f-"V$RW!feQKRQKfeQVRQ$f-"+(M"
##=
"
V#RWKfeQQR>!feVWR#>f"@A'D*G*0(eA'D+OOeC!5GD0A'"+B"#W5-"#>5-"+(M"$W5C0OM"(0A6+O"D0F'A+,'R"7&'"
##>
"
(/6;'AG"P'A'"KQR$KfeV>RKVfeLVRL>f-"LWR#!feVKR!WfeQ!R!=f-"+(M"LWR>KfeVKR=$feQ!R=Vf"
##K
"
*("B&'"@A*0A*Yk'5A'D+OO"60M'R"
##L
"
"
##Q
"
_*,/A'"=;"G&0PG"B&'"@'AC0A6+(D'"0C"3O+*A1"*("@A*0A*Yk'5A'D+OO"60M'-";A08'("M0P("B0"C0/A"]^_"
##V
"
A+(,'GT"WR>5!-"WR#5WR>-"WR!5WR#-"+(M"WRW>5WR!R"^B"M*m'A'(B"D0F'A+,'G-"B&'"@'AC0A6+(D'"0C"3O+*A1"+B"
#$W
"
A+(,'"WR#5WR>"\O0P56*M`"P+G"C0/(M"B0";'"+G",00M"+G"WR>5!"\6*M5&*,&`R"_0A"')+6@O'-"+B">We#>5C0OM"
#$!
"
B/60Ae(0A6+O"D0F'A+,'-"3O+*A1"+D&*'F'M"+"V=RLfeVVR$feVKRVf"@A'D*G*0(eA'D+OOeC!5GD0A'"+B"WR>5
#$#
"
!-"+(M"V>R#feVQR!feVKRLf"+B"WR#5WR>R"^B"A+(,'"WR!5WR#-"@A'D*G*0("P+G"A'M/D'M"B0"xKWf-"P&*O'"
#$$
"
A'D+OOG"@O+B'+/'M"+B"xVWfR"^B"A+(,'"WRW>5WR!-"@A'D*G*0("P+G"C/AB&'A"A'M/D'M"B0";'O0P"!Wf"
#$=
"
\!!RQf-">RLf-"+(M"=RKf"+B"#>5-">W5-"+(M"L>5C0OM"B/60A"D0F'A+,'`-"P&*O'"A'D+OOG"P'A'"A+*G'M"P*B&"
#$>
"
*(DA'+G*(,"B/60A"D0F'A+,'"\!KR>f-"$#RQf-"+(M"=KR!f`R"7&'"A'+G0("C0A"B&'"MA0@"*("@A'D*G*0("P+G"
#$K
"
B&+B"B&'"&*,&'A"D0F'A+,'"O'M"B0"+"MA+GYD"*(DA'+G'"*("B&'"(/6;'A"0C"F+A*+(B"D+(M*M+B'G"+B"F'AH"
#$L
"
O0P"]^_R"^B"A+(,'"WRW>5WR!-"B&'"(/6;'A"0C"D+(M*M+B'G"P+G"+;0/B"!$!8-"$!W8-"+(M"=!V8"+B"#>5-"
#$Q
"
>W5-"+(M"L>5C0OM"B/60A"D0F'A+,'R"
#$V
"
"
#=W
"
!"#$%#&'()"*'-*/,0"#"(-*-1&%#*91#,3"4*'(/*(%#&'5*)%(-'&,('3%(8"9'"+GG'GG'M"B&'"
#=!
"
@'AC0A6+(D'"0C"3O+*A1"+B"M*m'A'(B"D06;*(+Y0(G"0C"B/60A"@/A*BH"\!RW-"WRQ-"WRK-"WR=-"+(M"WR#`"+(M"
#=#
"
(0A6+O"@/A*BH"\!RW-"WRV>-"+(M"WRVW`R"^OO"B&'"')@'A*6'(BG"*("B&*G"G'DY0("/G'M">W5C0OM"B/60A"+(M"
#=$
"
#>5C0OM"(0A6+O"D0F'A+,'R"7&'"A'G/OBG"+A'"G&0P("*("_*,/A'"=D"+(M"1/@@O'6'(B+AH"7+;O'"=R"7&'"
#==
"
BP0"60M'G"\@A*0A*Yk'5C!"+(M"@A*0A*Yk'5A'D+OO`";'&+F'M"M*m'A'(BOH"P*B&"F+AH*(,"@/A*BHR"9*B&"
#=>
"
(0A6+O"@/A*BH"[)'M"+B"!RW-"*("@A*0A*Yk'5C!"60M'-"@A'D*G*0("A'6+*('M"+;0F'"VWf"\V$RW!f-"
#=K
"
VKR#>f-"VLRLVf-"VQRKQf-"+(M"VQRVVf"+B"B/60A"@/A*BH"!RW-"WRQ-"WRK-"WR=-"+(M"WR#`-"P&*O'"A'D+OO"
#=L
"
MA0@@'M"\QKRQKf-"Q!RK$f-"L!RWQf-">#RV=f-"+(M"##R=$f`"P*B&"M'DA'+G*(,"B/60A"@/A*BHR"X("B&'"
#=Q
"
@A*0A*Yk'5A'D+OO"60M'O-"@A'D*G*0("F+A*'M"\LWR#!f-"QWR!!f-"QQR!=f-"V$RQ!f-"+(M"VLR>=f`-"P&*O'"
#=V
"
A'D+OO"P+G";00GB'M-"'G@'D*+OOH"+B"O0P'A"B/60A"@/A*BH"\VKR!Wf-"V=R=>f-"VWR>!f-"QWRQ!f-"+(M"
#>W
"
>$RWKf`R"^DD0AM*(,"B0"B&'G'"A'G/OBG-",'('A+OOH-"P'"G/,,'GB"/G*(,"@A*0A*Yk'5C!"60M'"+B"&*,&'A"
#>!
"
B/60A"@/A*BH"+(M"@A*0A*Yk'5A'D+OO"60M'"+B"O0P'A"B/60A"@/A*BHR""
#>#
"
"
#>$
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
i/A"A'G/OBG"+OG0"G&0P'M"B&+B"O0P'A"(0A6+O"@/A*BH"&+A6'M"G06+YD"F+A*+(B5D+OO*(,"@'AC0A6+(D'-"
#>=
"
'G@'D*+OOH"0("A'D+OOR"9*B&"B/60A"@/A*BH"[)'M"+B"!RW-"*("@A*0A*Yk'5C!"60M'-"B&'"A'D+OOG"P'A'"
#>>
"
QKRQKf-"LWR$>f-"+(M">!RL=f"+B"(0A6+O"@/A*BH"!RW-"WRV>-"+(M"WRVWR"X("@A*0A*Yk'5A'D+OO"60M'-"B&'"
#>K
"
A'D+OOG"P'A'"VKR!Wf-"Q>R$Kf-"+(M"KVR=#fR"^DD0AM*(,OH-"P'"G/,,'GB"/G*(,"+"&*,&5@/A*BH"(0A6+O"
#>L
"
G+6@O'"P*B&"3O+*A1"C0A"G06+YD"F+A*+(B"M*GD0F'AHR"2+GBOH-"P'"G&0P'M"B&'"A'G/OBG"/G*(,"WRQeWRV>"
#>Q
"
B/60Ae(0A6+O"@/A*BHR"X("@A*0A*Yk'5A'D+OO"60M'-"3O+*A1"+D&*'F'M"+"LVRKQfeQ!R!$feQWR=Wf"
#>V
"
@A'D*G*0(eA'D+OOeC!5GD0A'-"M'60(GBA+Y(,"3O+*A1n"A'O*+;*O*BH"*("D&+OO'(,*(,"G+6@O'"D0(M*Y0(GR"
#KW
"
"
#K!
"
:('5;4,4*%$*<'54"*!%4,36"*'(/*<'54"*="7'36"*)'5548"EG*(,">W5C0OM"B/60A"+(M"#>5C0OM"(0A6+O"
#K#
"
D0F'A+,'-"P'"6+(/+OOH"+(+OHk'M"$WW"C+OG'"@0G*YF'"+(M"$WW"C+OG'"(',+YF'"D+OOG"A+(M06OH"@*D8'M"
#K$
"
CA06"+OO"F+A*+(B"D+OOGR"S+D&"_q"+(M"_a"P+G"+GG*,('M"P*B&"B&'"60GB"0;F*0/G"O*6*B+Y0("+G"B&'"
#K=
"
A'+G0("P&H"B&'"D+OO"P+G"C+OG'R"7&'"A'+G0(G"C0A"B&'"KWW"C+OG'"D+OOG"+A'"O*GB'M"*("1/@@O'6'(B+AH"
#K>
"
7+;O'">R"^"M*GBA*;/Y0("0C"B&'"A'+G0(G"*G",*F'("*("_*,/A'"="+G"+"@*'"D&+ABR"
#KK
"
"
#KL
"
^60(,"B&'"C+OG'"@0G*YF'"D+OOG-"$Vf"&+M"(0"6+BD&*(,"BA/B&";/B"P'A'"P*B&"B/60A"WRW>z]^_{WR!-"
#KQ
"
##f"P'A'"P*B&"B/60A"WR!z]^_{WR!>-"+(M"Vf"P'A'"P*B&"B/60A"]^_uWR!>R"^G"B&'"h'B&0M"G'DY0("
#KV
"
'O+;0A+B'G-"B&'G'"D+OOG"+A'"P*B&"B/60A"+(M"(0A6+O"D0F'A+,'"u=-"+(M"(0A6+O"]^_";'O0P"WRW>R"
#LW
"
i('"@0GG*;O'"')@O+(+Y0("*G"B&+B"3O+*A1"P+G"/(+;O'"B0"@*D8"/@"60A'"&*(BG"+(M"B'OO"B&'G'"D+OOG"CA06"
#L!
"
B&'"BA/'"0('GR"106'"0C"B&'G'"D+G'G"6*,&B";'"D0AA'DBOH"D+OO'M"+,+*("P*B&"&*,&'A"B/60A"D0F'A+,'-"
#L#
"
P&*D&"A'M/D'G"B&'"GB+YGYD+O";*+G"*("]^_R"XB"*G"+OG0"@0GG*;O'"B&+B"G*(D'"B&'"1Sb3#"BA/B&"G'B"*G"GYOO"
#L$
"
/(M'A"+DYF'"M'F'O0@6'(B-"*BG"*(D06@O'B'('GG"D+/G'M"+"C'P"F+A*+(BG"B&+B"+A'"+DB/+OOH"BA/'"B0";'"
#L=
"
6*GDO+GG*['M"+G"C+OG'"@0G*YF'GR"^(0B&'A"6+t0A"D+B',0AH"0C"C+OG'"@0G*YF'"D+OOG"*G"O*8'OH"B0";'"
#L>
"
D+/G'M";H"+O*,(6'(B"+AYC+DBG";'D+/G'"0C"+"A'@'YYF'"0A"*6@'AC'DB",'(06'"A'C'A'(D'"G'N/'(D'R"
#LK
"
XB"*(DO/M'G"Kf"vi('"0A"60A'"M'O'Y0(G"*("p+(8*(,">W;@w-">f"vX("O0P"D06@O')*BH"A',*0(w-"=f"
#LL
"
vS)D'GG*F'"6*G6+BD&'G"*("+O*,(6'(Bw-"$f"vi('"0A"60A'"*(G'AY0(G"*("p+(8*(,">W;@w-"+(M"*("
#LQ
"
B0B+O-"+(0B&'A"!Wf"*("A'@'YYF'"A',*0(G"0C"M*m'A'(B"BH@'GR"^OG0-"P'"C0/(M"!f"C+OG'"@0G*YF'"D+OOG-"
#LV
"
@0GG*;OH";'D+/G'"0C"*(G/|D*'(B"(0A6+O"D0F'A+,'R"
#QW
"
"
#Q!
"
^60(,"B&'"C+OG'"(',+YF'"D+OOG-"=Wf"BA/B&"F+A*+(BG"B&+B"P'A'"(0B"D+OO'M"P'A'"P*B&"B/60A"
#Q#
"
]^_{WR!-"!Wf"P'A'"P*B&"va0A6+O"]^_"uWRW>-";/B"B/60A"]^_"{K"Y6'G"O+A,'A"B&+("(0A6+O"]^_w-"
#Q$
"
+(M"$f"P'A'"P*B&"{$"A'+MG"G/@@0AY(,"B&'"F+A*+(B"+OO'O'"*("+"B/60AR"7&'G'"6*GG'M"F+A*+(BG"6*,&B"
#Q=
"
;'"D+OO'M"+,+*("*C"&*,&'A"B/60A"D0F'A+,'"*G",*F'(R"_+OG'"(',+YF'"D+OOG"+A'"60A'"O*8'OH"B0";'"
#Q>
"
D+/G'M";H"+O*,(6'(B"+AYC+DBG-"+G"P'"0;G'AF'M"!Wf"C+OG'"(',+YF'"D+OOG"*("B&'"&060@0OH6'A"
#QK
"
A',*0(-"!Wf"*("B&'"O0P"D06@O')*BH"A',*0(-"Lf"*("B&'"B+(M'6"A'@'+B"A',*0(-"+(M"*("B0B+O-"+(0B&'A"
#QL
"
Vf"B&+B"P'A'"+OG0"O*8'OH"B0"&+F'";''("D+/G'M";H"+O*,(6'(BR"9'"+OG0"0;G'AF'M"=f"C+OG'"(',+YF'"
#QQ
"
D+OOG"D+/G'M";H"')BA'6'"GBA+(M";*+G"\*R'R-"A'+MG"0;G'AF'M"*("0(OH"0('"GBA+(M`o"+(M"!f"P*B&";+G'"
#QV
"
N/+O*BH"0C"+OO"G/@@0AY(,";+G'G"{#WR"
#VW
"
"
#V!
"
>%&'3)*?(/"5@)'55,(7*9"#$%#&'()"8"7&'"1Sb3#"BA/B&"G'B"@A0F*M'G"x=W8"1a]G";/B"0(OH"x#8"X(M'OGR"
#V#
"
iP*(,"B0"B&'"GD+AD*BH"0C"BA/B&"G06+YD"X(M'OG"B&+B"D0/OM"'M/D'"GB+YGYD+O";*+G'G-"P'"
#V$
"
;'(D&6+A8'M"G06+YD"X(M'O5D+OO*(,"G'@+A+B'OHR"7&'"A'G/OBG"0C"M*m'A'(B"D06;*(+Y0(G"0C"B/60A"
#V=
"
D0F'A+,'"\#>5-">W5-"+(M"L>5C0OM`"+(M"(0A6+O"D0F'A+,'"\#W5-"#>5-"+(M"$W5C0OM`"+A'"G&0P("*("
#V>
"
1/@@O'6'(B+AH"7+;O'"KR"9*B&"(0A6+O"D0F'A+,'"[)'M"+B"#>5C0OM-"*("@A*0A*Yk'5C!"60M'"\F+A*+(B"
#VK
"
N/+O*BH"D/B0m"+B"!#`"3O+*A1"+D&*'F'M"LKRV!fe=VR$KfeKWR!$f-"KKR>=feKKRQVfeKKRL#f-"+(M"
#VL
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
>LRLVfeL!R=QfeK$RV!f"0C"@A'D*G*0(eA'D+OOeC!5GD0A'"+B"#>5-">W5-"+(M"L>5C0OM"B/60A"D0F'A+,'R"X("
#VQ
"
@A*0A*Yk'5A'D+OO"60M'"\D/B0m"+B"Q`-"3O+*A1"+D&*'F'M">$RW>feK>R=Kfe>QRK!f-"
#VV
"
$QRK$feLKR!>fe>!R#>f-"+(M"$LR>>feLKRKWfe>WR=WfR"^G"*("0/A"D0(DO/G*0("C0A"1a]"D+OO*(,-"
$WW
"
,'('A+OOH-"P'"G/,,'GB"/G*(,"@A*0A*Yk'5A'D+OO"60M'"C0A"D+OO*(,"G06+YD"X(M'OG"+B"O0P'A"B/60A"
$W!
"
D0F'A+,'R""
$W#
"
'
$W$
"
!"#$%#&'()"*'A"#*'//,(7*9.'4,(7*,($%#&'3%(*-%*-."*,(91-*,(*4-"9*B8"7&'"A'D0(GBA/DY0("0C"
$W=
"
&+@O0BH@'G-"+OG0"8(0P("+G"&+@O0BH@'5A'G0OF'M"+GG'6;OH"0A"@&+G*(,-",A'+BOH"*6@A0F'M"B&'"
$W>
"
@'AC0A6+(D'"0C"O0(,5A'+M",'A6O*('"F+A*+(B"D+OO*(,"*("@A'F*0/G"@A+DYD'22, 23R"X("B&'"h'B&0M"
$WK
"
G'DY0(-"P'"'O+;0A+B'"&0P"@&+G*(,"*(C0A6+Y0("D+(";'"/G'M"B0"'(&+(D'"G06+YD"F+A*+(B5D+OO*(,"
$WL
"
@'AC0A6+(D'R"EG*(,">We#>5C0OM"I33!$V>ed2"+(M"@A*0A*Yk'5A'D+OO"60M'"C0A";'(D&6+A8*(,-"P'"
$WQ
"
G&0P"*("S)B'(M'M"_*,/A'"!;"B&+B"@&+G+;O'"G06+YD"F+A*+(BG"@'AC0A6'M";'g'A"B&+("/(@&+G+;O'"
$WV
"
0('G-"'G@'D*+OOH"+B"O0P'A"]^_R"^B"]^_"WR!5WR!>-"@&+G+;O'"G06+YD"F+A*+(BG"&+M"+"
$!W
"
=$RQfeQ>RKfe>LRVf"@A'D*G*0(eA'D+OOeC!5GD0A'-";/B"/(@&+G+;O'"G06+YD"F+A*+(BG"&+M"0(OH"
$!!
"
!KR=feQ!R#fe#LR$fR"^B"]^_"WRW>5WR!-"@&+G+;O'"G06+YD"F+A*+(BG"&+M">RQfe=KRKfe!WR$f-";/B"
$!#
"
/(@&+G+;O'"G06+YD"F+A*+(BG"&+M"0(OH"#R$fe#LR#fe=R$fR"9'"BA*'M"M*G+;O*(,"@&+G*(,"*("3O+*A1"
$!$
"
+(M"C'M"(0"@&+G*(,"*(C0A6+Y0("B0"B&'"D+OO*(,"('BP0A8GR"^G"G&0P("*("1/@@O'6'(B+AH"7+;O'"L-"B&'"
$!=
"
0F'A+OO"_!5GD0A'"MA0@@'M"CA06"Q!R!=f"B0"LQRKKf"\5#R=Qf`R*
$!>
"
'
$!K
"
!"#$%#&'()"*%$*-."*-+%*#"49")36"*("-+%#C4*,(*4-"9*B8"X("D0(BA+GB"B0"3O+*A$-"*("P&*D&"B&'"@*O'/@"
$!L
"
('BP0A8"&+(MO'G"+OO"B&'"F+A*+(B"D+(M*M+B'G-"+(M"B&'"C/OO5+O*,(6'(B"('BP0A8"@A0D'GG'G"0(OH"B&'"
$!Q
"
/(M'D*M'M"D+(M*M+B'G"/G*(,"B&'"@*O'/@"('BP0A8-"3O+*A1"/G'G";0B&"('BP0A8G"'N/+OOH"B0"6+8'"
$!V
"
D0OO'DYF'"M'D*G*0(GR"7&'"A+Y0(+O'"+(M"M'B+*OG"+A'"'O+;0A+B'M"*("B&'"h'B&0M"G'DY0(R"9'"B'GB'M"
$#W
"
B&'"@'AC0A6+(D'"/G*(,";0B&"B&'"@*O'/@"('BP0A8"0(OH"+(M"B&'"C/OO5+O*,(6'(B"('BP0A8"0(OH-"P*B&"
$#!
"
>We#>5C0OM"0C"I33!$V>ed2"+(M"@A*0A*Yk'5A'D+OO"60M'R"7&'"A'G/OBG"+A'"G&0P("*("1/@@O'6'(B+AH"
$##
"
7+;O'"LR"9&'("P'"/G'M"0(OH"B&'"@*O'/@"('BP0A8-"B&'"_!5GD0A'"MA0@@'M"CA06"Q!R!=f"B0"L$R#Lf"
$#$
"
\5LRQLf`R"9&'("P'"/G'M"0(OH"B&'"C/OO5+O*,(6'(B"('BP0A8-"B&'"_!5GD0A'"MA0@@'M"CA06"Q!R!=f"B0"
$#=
"
LVR!=f"\5#RWWf`R*
$#>
"
"
$#K
"
!"#$%#&'()"*%$*>-"9DE*>"'#).,(7*$%#*'()"4-#'5*.'95%-;9"*4199%#-8"^G"'O+;0A+B'M"*("B&'"h'B&0M"
$#L
"
G'DY0(-"GB'@"$"/YO*k'G"A'60B'"+O*,(6'(B"G*,(+OG"B&+B"D0/OM"(0B";'"*(DO/M'M"*("B&'"('BP0A8"
$#Q
"
*(@/BG"M/'"B0"D06@/B+Y0(+O"O*6*B+Y0(G"B0"*6@A0F'"B&'"@A'D*G*0("0C"B&'"D+OO'M"G06+YD"F+A*+(BGR"
$#V
"
^G"G&0P("*("1/@@O'6'(B+AH"7+;O'"L-"P*B&0/B"B&*G"GB'@-"B&'"@A'D*G*0("MA0@@'M"CA06"LWR#!f"B0"
$$W
"
KLR!=f"\5$RWLf-"/G*(,">We#>5C0OM"I33!$V>ed2"+(M"@A*0A*Yk'5A'D+OO"60M'`R"
$$!
"
"
$$#
"
ClairS performance on Illumina data
$$$
"
1&0AB5A'+M"G06+YD"G6+OO5F+A*+(B"D+OO*(,"&+G";''("*(B'(G*F'OH"GB/M*'MR"^"(0(5')&+/GYF'"O*GB"0C"
$$=
"
GB+B'50C5B&'5+AB"6'B&0MG"*(DO/M'"1BA'O8+#10-"h/B'DB#8-"2+(D'B6-"a'/G06+YD5-"iDB0@/G12-"
$$>
"
106+YD1(*@'A9-"+(M"]+A('B4R"3O+*A1"P+G"M'G*,('M"@A*6+A*OH"C0A"O0(,5A'+M"G06+YD"G6+OO5F+A*+(B"
$$K
"
D+OO*(,R"I0P'F'A-"*BG"@0G'G"(0"O*6*B+Y0(G"B0"G6+OO"A'+MG-"+(M"P'"')@'DB"+"F+A*+(B5D+OO*(,"6'B&0M"
$$L
"
B&+B"P0A8G"C0A"O0(,"A'+MG"B0"@'AC0A6"+G"P'OO"+G"0A"'F'(";'g'A"B&+("')*GY(,"G&0AB5A'+M"6'B&0MGR"
$$Q
"
^OG0-";'(D&6+A8*(,"+,+*(GB"0B&'A"G&0AB5A'+M"G06+YD"G6+OO5F+A*+(B"D+OO'AG"@A0F*M'G"*(G*,&BG"0("
$$V
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
&0P"3O+*A1"@'AC0A6G"+,+*(GB"')*GY(,"6'B&0MG-"G*(D'"(0"0B&'A"O0(,5A'+M"D+OO'AG"+A'"+F+*O+;O'"C0A"
$=W
"
D06@+A*G0(R"
$=!
"
"
$=#
"
^"G/66+AH"0C"B&'"XOO/6*(+"M+B+"/G'M"C0A"60M'O"BA+*(*(,"+(M";'(D&6+A8*(,"*G"G&0P("*("
$=$
"
1/@@O'6'(B+AH"7+;O'"!R"9'"/G'M";0B&"IlWW$eIlWW="+(M"IlWW=eIlWW$"+G"B/60Ae(0A6+O"
$==
"
G+6@O'G"C0A"M+B+"GH(B&'G*GR"7&'"IlWW$"+(M"IlWW="G+6@O'G"&+M"V!R!>5"+(M"QQR=V5C0OM"D0F'A+,'-"
$=>
"
+(M"P'A'"@/;O*DOH"G&+A'M";H"l00,O'"I'+OB&"3'(B'AR"_0A";'(D&6+A8*(,-"P'"+OG0"/G'M"B&'"
$=K
"
I33!$V>ed2"B/60A5(0A6+O"@+*A-";/B"P*B&"M+B+"CA06"G*)"G'N/'(D*(,"D'(B'AG-"6+M'"+F+*O+;O'";H"
$=L
"
B&'"1Sb3#"D0(G0AY/6"s"a1T"a0F+1'N"+B"XOO/6*(+o"a3T"I*1'N"+B"B&'"a+Y0(+O"3+(D'A"X(GYB/B'-"X2o"
$=Q
"
I*1'N"+B"XO/6*(+-"S^o"I*1'N"+B"S/A0@'+("X(CA+GBA/DB/A'"C0A"7A+(GO+Y0(+O"h'M*D*('-"_?o"I*1'N"+B"
$=V
"
_/M+("E(*F'AG*BH-"a]o"I*1'N"+B"a0F+AYG"s"P*B&"D0F'A+,'"A+(,*(,"CA06"$LRV$5"B0"QLR>=5C0OMR"7&'"
$>W
"
G*)"6/OY5D'(B'A"A'@O*D+B'G"'(+;O'M"/G"B0"F'A*CH"3O+*A1n"@'AC0A6+(D'"D0(G*GB'(DHR"XC"(0B"
$>!
"
G@'D*[D+OOH"6'(Y0('M-"0B&'A"BA+*(*(,"+(M";'(D&6+A8*(,"M'B+*OG"+A'"B&'"G+6'"+G"B&0G'"0C"B&'"
$>#
"
ia7"M+B+"')@'A*6'(BG-"+(M"P'";'(D&6+A8'M"0(OH"1a]G"C0A"+OO"D+OO'AGR"2*8'"B&'"ia7"M+B+"
$>$
"
')@'A*6'(BG-"B&'"D066+(M"O*('G"+(M"O*(8G"B0"B&'"M+B+"/G'M"C0A";0B&"60M'O"BA+*(*(,"+(M"
$>=
"
;'(D&6+A8*(,"+A'"O*GB'M"*("B&'"1/@@O'6'(B+AH"a0B'GR"
$>>
"
"
$>K
"
7&'"'F+O/+Y0("6'BA*DG"0C"B&'"'*,&B"D+OO'AG"0("B&'"G*)"M+B+G'BG"+A'"G&0P("*("1/@@O'6'(B+AH"7+;O'"
$>L
"
QR"_*,/A'"K+"G&0PG"B&'"@A'D*G*0(5A'D+OO"D/AF'G-"+(M"[,/A'"K;"G&0PG"+"&*GB0,A+6"0C"B&'"_!5GD0A'GR"
$>Q
"
3O+*A1"D0(G*GB'(BOH"@'AC0A6'M"D06@+A+;OH"0A"GO*,&BOH";'g'A"B&+("B&'"BP0"B0@5@'AC0A6*(,"D+OO'AG-"
$>V
"
1BA'O8+#"+(M"h/B'DB#R"i("B&'"G*)"M+B+G'BG-"3O+*A1"+D&*'F'M"+"VLRQQf-"VLRQ#f-"VLR=Kf-"VLR>$f-"
$KW
"
VLRW$f-"+(M"VKR=!f"_!5GD0A'R"1BA'O8+#"+D&*'F'M"+"VKR!Kf-"VKR!Qf-"VLRW$f-"VLR$>f-"VKR$#f-"
$K!
"
+(M"V>R=LfR"h/B'DB#"+D&*'F'M"V>R#!f-"V>RL>f-"V>R$$f-"VKR#=f-"VKR=Kf-"+(M"V=R$>fR"dA08'("
$K#
"
M0P("*(B0"M*m'A'(B"]^_"A+(,'G-"B&'"3O+*A1"@'AC0A6+(D'"P+G"+OG0"D0(G*GB'(BOH"D06@+A+;O'"B0"0A"
$K$
"
;'g'A"B&+("B&+B"0C"0B&'A"D+OO'AG-"+G"G&0P("*("_*,/A'"KDR"_*,/A'"KM"G&0PG"]'(("M*+,A+6G"0C"B&'"
$K=
"
0F'AO+@G"0C"C+OG'"@0G*YF'"D+OOG";'BP''("1BA'O8+#-"h/B'DB#-"+(M"3O+*A1R"7&'"M*+,A+6G"G&0P"B&+B"
$K>
"
+OB&0/,&"A+A'-"B&'A'"+A'"$W"B0"=!"C+OG'"@0G*YF'"D+OOG"*("B&'"G*)"M+B+G'BG"B&+B"P'A'"D+OO'M";H"+OO"
$KK
"
B&A''"D+OO'AGR"iP*(,"B0"B&'"@0GG*;O'"*(D06@O'B'('GG"0C"B&'"BA/B&"G'B-"+"C+OG'"@0G*YF'"F+A*+(B"D+("
$KL
"
;'"'*B&'A"+"PA0(,"D+OO"0A"+"BA/'"F+A*+(B"6*GG*(,"CA06"B&'"BA/B&"G'BR"7&'G'"C+OG'"@0G*YF'"F+A*+(B"
$KQ
"
D+OOG"D+OO'M";H"+OO"B&A''"D+OO'AG"+A'"P0AB&"D0(M/DY(,"C/AB&'A"F'A*[D+Y0("+(M"6*,&B"C/AB&'A"
$KV
"
D0(BA*;/B'"B0"B&'"D06@O'B'('GG"0C"B&'"BA/B&"G'BR"
$LW
"
"
$L!
"
Discussion
$L#
"
X("B&*G"GB/MH-"P'"@A'G'(B"3O+*A1-"B&'"[AGB"G06+YD"G6+OO"F+A*+(B"D+OO'A"C0A"ia7"O0(,5A'+MGR"X("0/A"
$L$
"
;'(D&6+A8G-"P'"G&0P'M"B&+B"*B"*G"A'O*+;O'"+B"M*m'A'(B"G+6@O'"D0F'A+,'G-"B/60A"@/A*Y'G"+(M"
$L=
"
(0A6+O"D0(B+6*(+Y0(GR"9*B&"B&'"BA+*(*(,"M+B+"GH(B&'G*G"6'B&0M"P'"M'F*G'M-"3O+*A1"D+(";'"
$L>
"
BA+*('M"C0A"G06+YD"G6+OO5F+A*+(B"D+OO*(,"C0A"+(H"G'N/'(D*(,"@O+}0A6R"9'"M'60(GBA+B'M"B&+B"
$LK
"
3O+*A1"@'AC0A6'M"+G"P'OO"+G"0A"'F'(";'g'A"B&+("B&'"B0@5@'AC0A6*(,"G06+YD"F+A*+(B"D+OO'AG"C0A"
$LL
"
XOO/6*(+"G&0AB"A'+MGR"3O+*A1"MA+PG"0("*BG",'A6O*('"F+A*+(B"D+OO'A"@A'M'D'GG0AGn"')@'A*'(D'-"P&*O'"
$LQ
"
/G*(,"+"A'M'G*,('M"P0A8p0P-"('BP0A8"+AD&*B'DB/A'-"('BP0A8"0/B@/B-"+(M"@0GB5@A0D'GG*(,"
$LV
"
@A0D'M/A'"C0A"B&'"60A'"D&+OO'(,*(,"G06+YD"F+A*+(B5D+OO*(,"B+G8GR"
$QW
"
"
$Q!
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
7&'"/G'"0C"O0(,"A'+MG"C0A"G06+YD"1]"M*GD0F'AH"&+G"/(A+F'O'M"D06@O')"G06+YD"1]G"B&+B"P'A'"
$Q#
"
@A'F*0/G"&+6@'A'M";H"G&0AB"A'+MG28R"9*B&"B&'"/(@A'D'M'(B'M"@0P'A"0C"O0(,"A'+MG"B0"D0F'A"
$Q$
"
A'@'YYF'",'(06'"A',*0(G-"P'"')@'DB"B&'"/G'"0C"O0(,"A'+MG"C0A"G06+YD"G6+OO"F+A*+(B"D+OO*(,"B0"
$Q=
"
A'F'+O"60A'"G06+YD"F+A*+(BG"B&+B"P'A'"@A'F*0/GOH"*(+DD'GG*;O'";H"G&0AB"A'+MG-"+(M"O'+M"B0"+"
$Q>
"
;'g'A"/(M'AGB+(M*(,"0C"B&'"6/B+Y0(+O"@A0D'GG'G"+(M"C/(DY0(+O"D0(G'N/'(D'G"0C"B&'"G06+YD"
$QK
"
F+A*+(BG"*("M*m'A'(B"D+(D'A"BH@'GR"70"+OO0P"60A'"A'G'+AD&'AG"B0"+D&*'F'"B&'G'",0+OG-"3O+*A1"P+G"
$QL
"
*(DO/M'M"+G"B&'"G6+OO"F+A*+(B"D+OO'A"*("ia7nG"G06+YD"F+A*+(B5D+OO*(,"P0A8p0P29R"
$QQ
"
"
$QV
"
?'G@*B'"/G*(,"ia7nG"O+B'GB"b#Wc"M+B+-"B&'"_!5GD0A'"C0A"G06+YD"*(M'OG"P+G"0(OH"xKWfR"^OB&0/,&"
$VW
"
B&'"@'AC0A6+(D'"O008G";'g'A"P&'("D0(G*M'A*(,"0(OH"B&'"G06+YD"*(M'OG"*("B&'"D0M*(,"
$V!
"
G'N/'(D'G-"B&'"@A06*G'"0C";'g'A"P&0O'",'(06'"G06+YD"*(M'O5D+OO*(,"@'AC0A6+(D'"O*'G"*("B&'"
$V#
"
D0(Y(/0/G"+MF+(D'6'(B"0C"ia7nG"G'N/'(D*(,"D&'6*GBAH"+(M";+G'5D+OO*(,"+O,0A*B&6R"
$V$
"
"
$V=
"
Method
$V>
"
Training data synthesis
$VK
"
F"("#'3(7*4;(-."3)*-1&%#4*'(/*4;(-."3)*(%#&'58"^"B/60A"D06@A*G'G"(0A6+O"D'OOG"+(M"B/60A"
$VL
"
D'OOGo"B&'"O+g'A"+A'"A',+AM'M"+G"C0A'*,(R"1*6*O+AOH-"B&'"(0A6+O"D'OOG"0C"+("*(M*F*M/+O"+A'"
$VQ
"
D0(G*M'A'M"C0A'*,("B0"B&'"(0A6+O"D'OOG"0C"+(0B&'A"*(M*F*M/+O-"+(M"F*D'"F'AG+R"7&'",'A6O*('"
$VV
"
F+A*+(BG"/(*N/'"B0"+("*(M*F*M/+O"D+("6*6*D"+"G06+YD"F+A*+(B"P&'("6*)'M"P*B&"+(0B&'A"
=WW
"
*(M*F*M/+OR"9*B&"*(G/|D*'(B"8(0P("BA/B&"G06+YD"F+A*+(BG"+(M"GB+(M+AM"B/60A5(0A6+O"G+6@O'"
=W!
"
@+*AG"+F+*O+;O'-"B&*G"0;G'AF+Y0("O+HG"B&'"C0/(M+Y0("C0A",'('A+Y(,"+6@O'"GH(B&'YD"G06+YD"
=W#
"
F+A*+(BG"CA06"8(0P("BA/B&",'A6O*('"F+A*+(BG"*("B&'"lX^d"A'C'A'(D'"G+6@O'G"P*B&"A'+O"
=W$
"
G'N/'(D*(,"M+B+"C0A"M''@5O'+A(*(,"60M'O"BA+*(*(,R"7&'"M'B+*O'M"P0A8p0P"*G"G&0P("*("_*,/A'"!+R"
=W=
"
EG*(,"QW5C0OM"lX^d"IlWW#"0C"ia7"9l1"+O*,(6'(BG"+G"B&'"G0/AD'"0C"B&'"B/60A"\&'A'+Z'A"
=W>
"
A'C'AA'M"B0"+G"^`"+(M">W5C0OM"0C"IlWW!"+G"B&'"G0/AD'"0C"(0A6+O"\d`"+G"+("')+6@O'-"P'"[AGB"G@O*B"
=WK
"
B&'"+O*,(6'(BG"0C";0B&"G+6@O'G"*(B0"G6+OO'A"D&/(8G-"'+D&"P*B&"=5C0OM"D0F'A+,'R"7&'"G6+OO'A"
=WL
"
D&/(8G"CA06";0B&"G+6@O'G"D+(";'"D06;*('M"B0"G*6/O+B'"!`"M*m'A'(B"+OO'O'"CA'N/'(D*'G-"G/D&"+G"
=WQ
"
D06;*(*(,"=W5C0OM"^-"*R'R-"!W"=5C0OM"D&/(8G"0C"^"P*B&"#W5C0OM"d-"*R'R-">"=5C0OM"D&/(8G"0C"d-"B0"
=WV
"
G*6/O+B'"+"GH(B&'YD"B/60A"G+6@O'"P*B&"+("*M'+O"KLf"+OO'O'"CA'N/'(DH-"*R'R-"=W5C0OM"0C"^"+,+*(GB"
=!W
"
KW5C0OM"0C"^cdo"#`"M*m'A'(B"D0F'A+,'"0C";0B&"B/60A"+(M"(0A6+O-"'R,R"*(DA'+G'"0A"M'DA'+G'"B&'"
=!!
"
(/6;'A"0C"D&/(8G"+G"(''M'Mo"+(M"$`"M*m'A'(B"O'F'OG"0C"D0(B+6*(+Y0("*("(0A6+O-"'R,R-"*(GB'+M"0C"
=!#
"
/G*(,"!WWf"^"+G"(0A6+O-"+MM*(,"0('"0A"60A'"D&/(8G"0C"d"*("(0A6+OR"
=!$
"
"
=!=
"
G%6"#'7"*'/6,)"*$%#*-."*4%1#)"*/'-'8"9'"+@@O*'M"BP0"A'GBA*DY0(G"B0"B/60A"+(M"(0A6+O"
=!>
"
GH(B&'G*GR"_*AGB-"B0"+F0*M"+(H";*+G'G"D+/G'M";H"A'/G*(,"*(M*F*M/+O"A'+MG-"P'"/G'M"G+6@O*(,"
=!K
"
P*B&0/B"A'@O+D'6'(B"*("B/60A"+(M"(0A6+O"GH(B&'G*GR"7&+B"*G-"+"D&/(8"B&+B"P+G"/G'M"D0/OM"(0B"
=!L
"
;'"/G'M"+,+*(R"1'D0(M-"P'"A'N/*A'M"B&'"GH(B&'YD"B/60A"B0"&+F'"'N/+O"0A"&*,&'A"D0F'A+,'"B&+("
=!Q
"
B&'"GH(B&'YD"(0A6+O"B0"+O*,("P*B&"D0660("@A+DYD'R"EG*(,"B&'"@A'F*0/G"')+6@O'-"*R'R-"QW5C0OM"^"
=!V
"
+G"B&'"G0/AD'"0C"+"B/60A-"+(M">W5C0OM"d"+G"B&'"G0/AD'"0C"(0A6+O-"*C"#W5C0OM"d"*G"A'G'AF'M"B0"6*)"
=#W
"
P*B&"^"C0A"B&'"B/60A-"B&'("$W5C0OM"d"*G"O'Z"C0A"(0A6+O-";/B"P'"D+("+D&*'F'"B/60A"@/A*BH"0(OH"
=#!
"
;'BP''("$$f"\!W5C0OM"^"c"#W5C0OM"d`"+(M"!WWf"\QW5C0OM"^"c"(0"d`R"I0P'F'A-"*C"$W5C0OM"d"*G"
=##
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
A'G'AF'M"B0"6*)"P*B&"^"C0A"B&'"B/60A-"B/60A"@/A*BH";'BP''("Wf"\(0"^"c"$W5C0OM"d`"+(M"!WWf"
=#$
"
\QW5C0OM"^"c"(0"d`"D+(";'"+D&*'F'M-";/B"0(OH"#W5C0OM"d"*G"O'Z"C0A"(0A6+OR"7&/G-"P'"G/,,'GB"/G*(,"
=#=
"
B&'"G+6@O'"P*B&"&*,&'A"D0F'A+,'"+G"B&'"G0/AD'"0C"(0A6+O"B0"+D&*'F'";0B&"&*,&'A"(0A6+O"
=#>
"
D0F'A+,'"+(M"C/OO5G@'DBA/6"B/60A"@/A*BH"C0A"60M'O"BA+*(*(,R"7&*G"*G"D0/(B'A*(B/*YF'";'D+/G'"
=#K
"
&*,&'A"B/60A"D0F'A+,'"0C"D0660("*("D+(D'A"GB/M*'GR"
=#L
"
"
=#Q
"
H"#,6,(7*&15395"*)'-"7%#,"4*%$*6'#,'(-4*$#%&*'*4;(-."3)*-1&%#I(%#&'5*9',#8"_0/A"D+B',0A*'G"0C"
=#V
"
F+A*+(BG-"(+6'OH"v106+YDw-"vl'A6O*('w-"v^AYC+DBw-"+(M"va0A6+O50(OHw-"P'A'"M'A*F'M"CA06"+"
=$W
"
GH(B&'YD"B/60Ae(0A6+O"@+*A-"+G"')@O+*('M"*("M'B+*O"*("_*,/A'"!+R"9'"/G'M"lX^d"BA/B&"F+A*+(BG"*("
=$!
"
B&'"GH(B&'YD"B/60A"+(M"GH(B&'YD"(0A6+O-"+(M"P'"M'[('M"+(H"BA/B&"F+A*+(BG"\*(DO/M*(,";0B&"
=$#
"
&060kH,0/G"+(M"&'B'A0kH,0/G`"+G"vD+OO+;O'5F+A*+(BwR"d+G*D+OOH-"G06+YD"F+A*+(BG"+A'"D+OO+;O'"
=$$
"
F+A*+(BG"*("B&'"GH(B&'YD"B/60A";/B"(0B"*("B&'"GH(B&'YD"(0A6+O-"P&*D&"+A'"+OG0"8(0P("+G"BA/B&"
=$=
"
,'A6O*('"F+A*+(BG"*("B&'"B/60A"G0/AD'-";/B"(0B"*("B&'"(0A6+O"G0/AD'R"l'A6O*('"F+A*+(BG"+A'"
=$>
"
D+OO+;O'"F+A*+(BG"*(";0B&"B&'"GH(B&'YD"B/60A"+(M"GH(B&'YD"(0A6+O-"P&*D&"+A'"+OG0"8(0P("+G"
=$K
"
BA/B&",'A6O*('"F+A*+(BG"*(";0B&"B&'"B/60A"G0/AD'"+(M"(0A6+O"G0/AD'R"^AYC+DB"F+A*+(BG"+A'"
=$L
"
D+OO+;O'"F+A*+(BG"0(OH"*("B&'"GH(B&'YD"B/60AR"7&'H"+A'"(0B"8(0P("+G"BA/B&"*("'*B&'A"B&'"B/60A"
=$Q
"
G0/AD'"0A"B&'"(0A6+O"G0/AD'R"a0A6+O50(OH"F+A*+(BG"+A'"D+OO+;O'5F+A*+(BG"C0/(M"0(OH"*("B&'"
=$V
"
GH(B&'YD"(0A6+OR"7&A''"0C"B&'"C0/A"D+B',0A*'G-"106+YD-"l'A6O*('"+(M"^AYC+DB-"+A'"/G'M"C0A"
==W
"
60M'O"BA+*(*(,R"i(OH"B&'"F+A*+(BG"O0D+B'M"*("B&'"0F'AO+@@*(,"lX^d5M'[('M"&*,&5D0([M'(D'"
==!
"
A',*0(G"0C";0B&"B&'"B/60A"+(M"(0A6+O"G0/AD'G"P'A'"/G'M"C0A"BA+*(*(,"B0"'(G/A'"B&'"N/+O*BH"0C"
==#
"
B&'"BA+*(*(,"F+A*+(BGR"d'D+/G'"0C"B&'"G/;5G+6@O*(,"@A0D'GG-"G06'"F+A*+(BG"6*,&B"&+F'"C'P"
==$
"
G/@@0AY(,"A'+MG"*("B&'"GH(B&'YD"B/60A-"'G@'D*+OOH"C0A"O0P"^_"G06+YD"F+A*+(BGR"7&'G'"F+A*+(BG"
===
"
P'A'"')DO/M'M"CA06"60M'O"BA+*(*(,"B0"+F0*M"D0(C/G*(,"B&'"('/A+O"('BP0A8R"h0A'"')DO/G*0("
==>
"
M'B+*OG"+A'",*F'("*("B&'"C0OO0P*(,"@+A+,A+@&R"i/A"G+6@O'"GH(B&'G*G"6'B&0M"G/@@0ABG",'('A+Y(,"
==K
"
GH(B&'YD"B/60AG"+B"+(H"@/A*BH"O'F'O-"G0"P'"D+("/G'"+G"6+(H"@/A*Y'G"+G"@0GG*;O'"B0"+D&*'F'"[('"
==L
"
D0F'A+,'"0C"]^_"CA06"W"B0"!-";/B"C0A"@A+DYD+O*BH-"P'"/G'M"B&A''"B/60A"@/A*Y'G"\#>f-">Wf-"+(M"
==Q
"
!WWf`-"+(M"+@@O*'M"G/;G+6@O*(,"B0"+OO"F+A*+(BG"CA06"B&'"B&A''"@/A*Y'G"B0"+D&*'F'"+DD'@B+;O'"
==V
"
]^_"M*GBA*;/Y0(R"7&*G"*G"C'+G*;O'";'D+/G'"!`"B&'"*((+B'"F+A*+(D'"0C"B&'"^_"0C"B&'",'A6O*('"
=>W
"
F+A*+(BG"CA06"B&'"B/60A"+(M"(0A6+O"G0/AD'G"'(+;O'G"+"@00O"0C"G06+YD"F+A*+(BG"C/OOH"D0F'A*(,"
=>!
"
]^_"CA06"W"B0"!-"'F'("P*B&"t/GB"B&A''"@/A*Y'G-"+(M"#`"+@@OH*(,"G/;G+6@O*(,"B0"B&'"@00O"'(+;O'G"
=>#
"
/G"B0"'(A*D&"M*|D/OB"G06+YD"F+A*+(BG"+(M"A'M/D'"B&'"(/6;'A"0C"O'GG"D0660("G06+YD"F+A*+(BG"
=>$
"
C0A"60M'O"BA+*(*(,R"X("B'A6G"0C"G/;G+6@O*(,-"B&'"]^_G"0C"D&0G'("G06+YD"F+A*+(BG"P'A'"A+(M06OH"
=>=
"
G'O'DB'M"CA06"+";'B+"M*GBA*;/Y0("P*B&"G&+@'"@+A+6'B'AG"~•#"+(M"€•>R"7&'"G+6'"M*GBA*;/Y0("
=>>
"
P+G"/G'M";H"B&'"1Sb3#"D0(G0AY/6"C0A"G@*8'5*("G06+YD"F+A*+(BG"*("1+&A+'*+("'B"+O30R"i/A"
=>K
"
')@'A*6'(B"G&0P'M"B&+B"G/;G+6@O*(,"*BG'OC"A'G/OB'M"*("+"x!RQf"*(DA'+G'"*("B&'"_!5GD0A'"/G*(,"
=>L
"
>W)e#>)"0C"I33!$V>ed2R"7&'"A'G/OY(,"]^_"M*GBA*;/Y0("0C"1a]G"*G"G&0P("*("_*,/A'"!;R"9'"BA*'M"
=>Q
"
+MM*(,"0('"60A'"B/60A"@/A*BH"+B"!#R>f-";/B"+@+AB"CA06"O0(,'A"60M'O"BA+*(*(,"Y6'-"(0"
=>V
"
@'AC0A6+(D'",+*("P+G"0;G'AF'MR""
=KW
"
"
=K!
"
J-."#*/"-',54*'2%1-*-."*6'#,'(-4*4"5")-"/*$%#*&%/"5*-#',(,(78"_0A"B&'"G06+YD"F+A*+(BG"/G'M"C0A"
=K#
"
60M'O"BA+*(*(,-"+"6*(*6/6"D0F'A+,'"0C"C0/A-"+(M"+"6*(*6/6"0C"B&A''"A'+MG"G/@@0AY(,"B&'"
=K$
"
G06+YD"F+A*+(B"+OO'O'"+A'"A'N/*A'MR"106+YD"F+A*+(BG"P*B&"]^_"•"WRW$"*("B&'"GH(B&'YD"(0A6+O"
=K=
"
P'A'"')DO/M'M"CA06"BA+*(*(,"B0"+F0*M"D0(C/G*(,"B&'"60M'O"P*B&"+"F'AH"(0*GH"(0A6+OR"_0A"B&'"
=K>
"
+AYC+DBG-"B&'"(0(5A'C'A'(D'"^_"P+G"D+@@'M"+B"WRW>"B0"+F0*M"/G*(,"+"O+A,'"(/6;'A"0C"0;F*0/G"
=KK
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
+AYC+DBG"C0A"BA+*(*(,R"_0A",'A6O*('"F+A*+(BG-"6*(*6/6"D0F'A+,'"0C"C0/A"A'+MG"+(M"+"6*(*6/6"0C"
=KL
"
B&A''"A'+MG"G/@@0AY(,"B&'",'A6O*('"F+A*+(B"+OO'O'-"P'A'"A'N/*A'M"*(";0B&"GH(B&'YD"B/60A"+(M"
=KQ
"
(0A6+OR"l'A6O*('"F+A*+(BG"P*B&"+"M*m'A'(D'"*("^_"O+A,'A"B&+("WR!";'BP''("B&'"GH(B&'YD"B/60A"
=KV
"
+(M"(0A6+O"P'A'"')DO/M'M"CA06"BA+*(*(,R"70 " @A'F'(B"B&'"60M'O"CA06"*(C'AA*(,"+"G06+YD"F+A*+(B"
=LW
"
CA06"*BG"D05')*GB'(D'"P*B&"BP0"0A"60A'"+Mt+D'(B",'A6O*('"F+A*+(BG-"P&*D&"*G"+"D0(C0/(M*(,"
=L!
"
C+DB0A"B&+B"D+(";'"'+G*OH"O'+A('M";H"B&'"60M'O-",'A6O*('G"F+A*+(BG"B&+B"P'A'"O'GG"B&+("$$;@"\B&'"
=L#
"
P*(M0P"G*k'"0C"0/A"60M'O"M'G*,(`"CA06"'+D&"0B&'A"P'A'"')DO/M'M"CA06"BA+*(*(,R"i/A"')@'A*6'(B"
=L$
"
G&0P'M"B&+B"B&*G"')DO/G*0("+O0('"*(DA'+G'M"G06+YD"F+A*+(B"D+OO*(,"@A'D*G*0(";H"x!fR"9*B&"B&A''"
=L=
"
B/60A"@/A*Y'G"+(M"B&'"')DO/G*0(G"')@O+*('M"+;0F'-"!#-=QV-$=#"BA+*(*(,"G+6@O'G"P'A'"O'ZR"7&'"
=L>
"
;A'+8M0P("*G"G&0P("*("_*,/A'"!DR""
=LK
"
"
=LL
"
!.'4,(7*,($%#&'3%(*"(.'()"4*4%&'3)*6'#,'(-@)'55,(7*9"#$%#&'()"8"^("+/B&'(YD"G06+YD"
=LQ
"
F+A*+(B"/G/+OOH"0A*,*(+B'G"CA06"'*B&'A"B&'"6+B'A(+O"0A"@+B'A(+O"&+@O0BH@'G-"P&*O'"+"A+(M06"'AA0A"
=LV
"
/G/+OOH"&+G"+"C+*A"D&+(D'"&+@@'(*(,"*(";0B&"\S)B'(M'M"?+B+"_*,/A'"!+`R"7&/G-"G06+YD"F+A*+(BG"
=QW
"
B&+B"&+F'"+"G*(,O'"+(D'GBA+O"&+@O0BH@'"\'*B&'A"6+B'A(+O"0A"@+B'A(+O`"G&0/OM";'"D0(G*M'A'M"60A'"
=Q!
"
A'O*+;O'"B&+("B&0G'"P*B&"BP0"+(D'GBA+O"&+@O0BH@'G-"')D'@B"C0A"G06+YD"F+A*+(BG"P*B&"&*,&"]^_"
=Q#
"
B&+B"6*,&B";'"+"A'G/OB"0C"D0@H"(/6;'A"+OB'A+Y0("0A"DO0(+O"M/@O*D+Y0(31R"3O+*A1"/G'G"@&+G*(,"
=Q$
"
*(C0A6+Y0("C0A";0B&"60M'O"BA+*(*(,"+(M"*(C'A'(D'R"3O+*A$"+(M"20(,q&+G'"+A'"/G'M"C0A"@&+G*(,"
=Q=
"
+(M"A'+M"&+@O0B+,,*(,R"h0A'"M'B+*OG"+A'",*F'("*("B&'"v3O+*A1"*(@/B"+(M"0/B@/Bw"G'DY0(R"3O+*A1"
=Q>
"
/G'G"@&+G*(,"*(C0A6+Y0("M/A*(,"C/OO5+O*,(6'(B5;+G'M"F+A*+(B"D+OO*(,-"*("P&*D&"+"D&+(('O"(+6'M"
=QK
"
v7/60Aea0A6+Oeq&+G*(,"X(C0w"*G"/G'MR"X("B&*G"D&+(('O-"B&'"+O*,(6'(BG"+A'",A0/@'M"*(B0"
=QL
"
&+@O0BH@'5/(8(0P(-"&+@O0BH@'"!-"+(M"&+@O0BH@'"#-"'+D&"/G*(,"B&'"A'+M"0AM'A"0C"B&'"
=QQ
"
+O*,(6'(BGR"^OB&0/,&"O0(,5A'+M"G'N/'(D*(,"'(+;O'G"0/BGB+(M*(,"@&+G*(,"@'AC0A6+(D'-"G06'"
=QV
"
G06+YD"F+A*+(BG"*("M*|D/OB",'(06*D"A',*0(G"0A"P*B&0/B"+"&'B'A0kH,0/G",'A6O*('"*("B&'*A"F*D*(*BH"
=VW
"
GYOO"D+((0B";'"D0F'A'M";H"+(H"@&+G'M"A'+MGR"7&/G-"M/A*(,"60M'O"BA+*(*(,-"C0A"'+D&"F+A*+(B"B&+B"
=V!
"
&+G"+"&'B'A0kH,0/G"0A*,*("CA06"B&'"B/60A"G0/AD'-"*C"0('"0A"60A'"A'+MG"D+(";'"@&+G'M-";0B&"+"
=V#
"
F'AG*0("0C"*(@/B"P*B&"A'+MG"+Z'A"@&+G*(,"+(M"+"F'AG*0(";'C0A'"@&+G*(,"P'A'"/G'MR"
=V$
"
"
=V=
"
ClairS workflow and design
=V>
"
J6"#6,"+8"_*,/A'"#"G&0PG"+("0F'AF*'P"0C"B&'"3O+*A1"G06+YD"F+A*+(B5D+OO*(,"P0A8p0PR"1B+AY(,"
=VK
"
CA06"B&'"+O*,(6'(BG"*("B&'"d^he3:^h"C0A6+B"0C"+"B/60Ae(0A6+O"G+6@O'"@+*A-"3O+*A1"C0OO0PG"
=VL
"
B&A''"GB'@G"B0"M'A*F'"B&'"G06+YD"F+A*+(BG"*("+"B/60A"+(M"0/B@/BG"B&'6"B0"+"]3_"[O'R"X("GB'@"!-"
=VQ
"
3O+*A1"/G'G"3O+*A$"+(M"20(,q&+G'"C0A",'A6O*('"F+A*+(B"D+OO*(,-"@&+G*(,"+(M"A'+M"&+@O0B+,,*(,R"
=VV
"
7&'"@A0D'GG'M"+O*,(6'(BG"+A'"B&'("/G'M"C0A";0B&"@*O'/@"+(M"C/OO5+O*,(6'(B5;+G'M"G06+YD"
>WW
"
F+A*+(B"D+OO*(,"*("GB'@"#R"1B'@"$"*(F0OF'G"@0GB5@A0D'GG*(,"[OB'AG"B&+B"'O*6*(+B'"G06+YD"F+A*+(B"
>W!
"
D+OO*(,"*C"+("+(D'GBA+O"&+@O0BH@'"\'*B&'A"6+B'A(+O"0A"@+B'A(+O`"CA06"P&*D&"B&'"G06+YD"F+A*+(B"
>W#
"
D0/OM"0A*,*(+B'"D+((0B";'"C0/(MR"
>W$
"
"
>W=
"
>-"9*KE*F"#&5,("*6'#,'(-*)'55,(7L*9.'4,(7*'(/*#"'/*.'95%-'77,(78"1B'@"!"*G"M'@*DB'M"*("_*,/A'"#;R"
>W>
"
3O+*A$22"*G"*(B',A+B'M"*(B0"3O+*A1"C0A"D+OO*(,"&*,&5N/+O*BH"&'B'A0kH,0/G",'A6O*('"F+A*+(BG"*(";0B&"
>WK
"
B/60A"+(M"(0A6+O"B0"6+)*6*k'"B&'"@'AC0A6+(D'"0C"B&'"G/;G'N/'(B"@&+G*(,"B+G8R"E(O*8'"3O+*A$nG"
>WL
"
M'C+/OB-"^_uWR#"+(M"D0F'A+,'u!W"P'A'"+@@O*'M"B0"'(G/A'"B&'"N/+O*BH"0C"B&'"D+OO'M"F+A*+(BG"+(M"
>WQ
"
A'M/D'"D06@/B+Y0(+O"0F'A&'+MR"i(OH"B&'"&'B'A0kH,0/G",'A6O*('"F+A*+(BG"C0/(M"*(";0B&"B/60A"
>WV
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
+(M"(0A6+O"P'A'"D&0G'("C0A"@&+G*(,R"_0A"@&+G*(,"+(M"&+@O0B+,,*(,"B&'"B/60A"+O*,(6'(BG-";0B&"
>!W
"
20(,q&+G'32"+(M"9&+BGI+@33+A'"+OO0P'M"*("3O+*A1R"9'"D&0G'"20(,q&+G'"0F'A"9&+BGI+@"+G"B&'"
>!!
"
M'C+/OB";'D+/G'"20(,q&+G'"A/(G"x!>"Y6'G"C+GB'A"P&*O'"M'O*F'A*(,"G*6*O+A"0A"O0(,'A"@&+G'"G'BG"
>!#
"
0("&/6+("G+6@O'GR"a0B+;OH-"3O+*A1"M0'G"(0B"@&+G'"+(M"&+@O0B+,"B&'"(0A6+O"+O*,(6'(BGR"i/A"
>!$
"
')@'A*6'(B"G&0P'M"@&+G*(,"B&'"(0A6+O"+O*,(6'(BG"M0/;O'M"B&'"@A0D'GG*(,"Y6'";/B"M*M"(0B"
>!=
"
A'G/OB"*("+(H"*6@A0F'6'(B"*("D+OO*(,"@'AC0A6+(D'R"
>!>
"
"
>!K
"
>-"9*B8*!,5"19@2'4"/*'(/*$155@'5,7(&"(-@2'4"/*6'#,'(-*)'55,(78"1B'@"#"*G"M'@*DB'M"*("_*,/A'"#DR"_0A"
>!L
"
+"F+A*+(B"D+(M*M+B'"\')@O+*('M"*("v1'O'DY(,"F+A*+(B"D+(M*M+B'Gw`-"+"@*O'/@"*(@/B"+(M"+"C/OO5
>!Q
"
+O*,(6'(B"*(@/B"+A'",'('A+B'M"\')@O+*('M"*("v7&'"M'G*,("0C"@*O'/@"*(@/B"+(M"C/OO5+O*,(6'(B"
>!V
"
*(@/Bw`R"7&'("B&'"*(@/BG"+A'"G'(B"B0"+"d*5l:E5;+G'M"@*O'/@5D+OO*(,"('/A+O"('BP0A8"+(M"+":'Ga'B5
>#W
"
;+G'M"C/OO5+O*,(6'(B5D+OO*(,"('/A+O"('BP0A8"\')@O+*('M"*("v7&'"M'G*,("0C"('/A+O"('BP0A8Gw`"C0A"
>#!
"
*(C'A'(D'R"d0B&"('BP0A8G"&+F'"B&'"G+6'"0/B@/B"s"+"G*(,O'"B+G8"P*B&"B&A''"D+B',0A*'G-"v106+YDw-"
>##
"
vl'A6O*('w-"+(M"v^AYC+DBw-"P&*D&"6+BD&"')+DBOH"B&'"B&A''"D+B',0A*'G"M'[('M"*("B&'"GH(B&'YD"
>#$
"
BA+*(*(,"M+B+R"X("D0(BA+GB"B0"3O+*A$-"*("P&*D&"B&'"C+GB'A"@*O'/@5;+G'M"D+OO*(,"DO'+(G"/@"60GB"
>#=
"
F+A*+(B"D+(M*M+B'G"B&+B"+A'"0;F*0/G"F+A*+(BG-"+(M"B&'"60A'"D06@/B+Y0(+O5M'6+(M*(,"C/OO5
>#>
"
+O*,(6'(B5;+G'M"D+OO*(,"&+(MO'G"B&'"BA*D8H"+(M"O'GG"0;F*0/G"D+(M*M+B'G-"3O+*A1"D0(G*M'AG"B&'"
>#K
"
@0P'A"0C"B&'"BP0"('/A+O"('BP0A8G"'N/+OR"9'"0;G'AF'M"B&+B"C/OO5+O*,(6'(B5;+G'M"D+OO*(,"*G"
>#L
"
@'AC0A6+(B"+B"6*M5A+(,'"]^_GR"I0P'F'A-"@*O'/@5;+G'M"D+OO*(,"A'N/*A'G"O'GG"'F*M'(D'"B&+("C/OO5
>#Q
"
+O*,(6'(B"D+OO*(,"B0"MA+P"B&'"G+6'"D0(DO/G*0(R"9&'("]^_",0'G"/(M'A"WR!-"@*O'/@5;+G'M"D+OO*(,"
>#V
"
;'D06'G"*(DA'+G*(,OH"60A'"G'(G*YF'"+(M"/G/+OOH"0/B@'AC0A6G"C/OO5+O*,(6'(B5;+G'M"D+OO*(,R"7&*G"
>$W
"
0;G'AF+Y0("6+8'G"@*O'/@5;+G'M"D+OO*(,"60A'"*6@0AB+(B"C0A"G06+YD"F+A*+(B"D+OO*(,"B&+("*BG"A0O'"
>$!
"
*("3O+*A$"C0A",'A6O*('"F+A*+(B"D+OO*(,-"'G@'D*+OOH"*("6/OY@O'"DO*(*D+O"/G+,'"GD'(+A*0G"P&'("
>$#
"
G'(G*YF*BH"*G"'6@&+G*k'MR"X("3O+*A1-"+"G06+YD"F+A*+(B"*G"D+OO'M"P&'(";0B&"('BP0A8G",*F'"G06+YD"
>$$
"
B&'"&*,&'GB"@A0;+;*O*BHR"7&'"F+A*+(B"N/+O*BH"\bE^2`"*G"q&A'M5O*8'"+(M"*G"D+OD/O+B'M"+G"
>$=
"
!"#$%&'()*+!"
,
!#$
$
-
./0(1
-"P&'A'"
23%!"#$%&'
(&)*+, &%!"#$%&'
-+)).$)&/0#*0%
'
R"^OG0"*("D0(BA+GB"B0"3O+*A$-"
>$>
"
P&*D&"/G'G"B&'"G+6'"('BP0A8"C0A";0B&"1aq"+(M"X(M'O"D+OO*(,-"3O+*A1"/G'G"BP0"M*m'A'(B"('BP0A8G"
>$K
"
A'G@'DYF'OH"BA+*('M"C0A"1a]"+(M"X(M'O"D+OO*(,R"7&*G"6'+(G"B&+B"3O+*A1"A/(G"0("C0/A"('BP0A8G"*("
>$L
"
B0B+OT"@*O'/@"C0A"1a]-"@*O'/@"C0A"X(M'O-"C/OO5+O*,(6'(B"C0A"1a]-"+(M"C/OO5+O*,(6'(B"C0A"X(M'OR"7&'"
>$Q
"
A+Y0(+O'";'&*(M"B&'"('P"M'G*,("*G"B&+B"/(O*8'",'A6O*('"F+A*+(BG"B&+B"+A'"D0660(OH"M*@O0*M-"
>$V
"
G06+YD"F+A*+(BG"&+F'"(0"@O0*MH"+GG/6@Y0(-"6'+(*(,"B&+B"B&'"')*GB'(D'"0C"1a]G"+(M"X(M'OG"*("
>=W
"
B&'"G+6'"@0G*Y0("+A'"*(M'@'(M'(B"'F'(BGR"i/A"B'GBG"C0/(M"B&+B"/G*(,"G'@+A+B'M"('BP0A8G"O'M"B0"
>=!
"
+"!R>f"*(DA'+G'"*("1a]"A'D+OOR"7&'"/G'"0C"G'@+A+B'M"('BP0A8G"+OG0"+OO0P'M"B&'"/G'"0C"M*m'A'(B"
>=#
"
F+A*+(B"N/+O*BH"D/B0mG"C0A"1a]"+(M"X(M'O-"P&*D&"*G"/G'C/O"C0A"G06+YD"F+A*+(B"D+OO*(,-"'G@'D*+OOH"
>=$
"
P&'("B&'"G+6@O'"D0(M*Y0("*G"(0B"*M'+OR"
>==
"
"
>=>
"
>"5")3(7*6'#,'(-*)'(/,/'-"48"1'(M*(,"'F'AH",'(06'"@0G*Y0("+G"+"F+A*+(B"D+(M*M+B'"B0"B&'"
>=K
"
('/A+O"('BP0A8G",/+A+(B''G"6+)*6/6"G'(G*YF*BHR"I0P'F'A-"*B"*G"(0B"0(OH"D06@/B+Y0(+OOH"
>=L
"
*(C'+G*;O'-";/B"+OG0"/(A'+G0(+;O'"B0"P0A8"0("(0(GB+AB'A"@0G*Y0(G-"G/D&"+G"B&0G'"P*B&0/B"+(H"
>=Q
"
(0(5A'C'A'(D'"+OO'O'"G/@@0ABR"^",00M"F+A*+(B"D+(M*M+B'"G'O'DY0("GBA+B',H"*G"'GG'(Y+O"B0"+D&*'F'"
>=V
"
+";+O+(D'";'BP''("G'(G*YF*BH"+(M"A/((*(,"Y6'R"X("3O+*A1-"B&'"G'O'DY0("DA*B'A*+"+A'"+G"C0OO0PGR"
>>W
"
2'B"#
4
M•\:L*GL*FL*N`";'"B&'"A'C'A'(D'";+G'"0C"+",'(06'"@0G*Y0(-"+(M"&
4
*M@#";'"B&'"+OB'A(+YF'"
>>!
"
;+G'GR"HXm"M'(0B'G"B&'"D0F'A+,'"0C"&"+B"B&'"@0G*Y0("*("G+6@O'"O
4
‚NL* = ƒ-"P&'A'"N"+(M"="
>>#
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
A'@A'G'(B"B&'"B/60A"+(M"(0A6+O"G+6@O'R"Gm"M'[('G"B&'"G'O'DY0("DA*B'A*+"0C"'+D&"+OB'A(+YF'"
>>$
"
;+G'"*("&-"+GT"
>>=
"
5(3
6
7
7
7
8
7
7
7
9
:;)<=$$$>?$@(
)$AB$*C$@(
)
D
@*
)+
*AE
FCG=$$$>?$@(
)
D
@*
)+
*HE$;IJ$@(
,
D
@*
,+
*AE
FCG=$$$>?$-#
1
∑
-&
12
&
3#
4
∑
3&
42
&
HK$;IJ$@(
,
D
@*
,+
*HE
"
>>>
"
P&'A'"~"G'BG"B&'"6*(*6/6"]^_-"+(M"€"G'BG"B&'"6*(*6/6"B/60A"]^_"B0"(0A6+O"]^_"A+Y0"C0A"+"
>>K
"
D+(M*M+B'"B0";'"G'O'DB'MR"X(B/*YF'OH-"B&'"[AGB"'N/+Y0("6'+(G"M*GA',+AM*(,"F+A*+(B"D+(M*M+B'G"
>>L
"
P*B&"{"$"A'+MG"*("B/60A"G/@@0AY(,"B&'"F+A*+(B"+OO'O'-"0A"P*B&"]^_"*("B/60A"{"~R"7&'"G'D0(M"
>>Q
"
'N/+Y0("6'+(G"G'O'DY(,"+"F+A*+(B"D+(M*M+B'"*C"*BG"]^_"*G"{~"*("(0A6+O-";/B"u~"*("B/60AR"7&'"
>>V
"
B&*AM"'N/+Y0("6'+(G"G'O'DY(,"+"F+A*+(B"D+(M*M+B'-"'F'("*C"*BG"]^_"*("(0A6+O"*G"u~-"B&'"]^_"*("
>KW
"
B/60A"*G"u€"Y6'G"O+A,'A"B&+("B&'"]^_"*("(0A6+OR"X("3O+*A1-"~"+(M"€"+A'"D0([,/A+;O'"+(M"M'C+/OB"
>K!
"
B0"WRW>"+(M"K-"A'G@'DYF'OHR"2*8'"60M'O"BA+*(*(,"M+B+"@A'@+A+Y0(-"D0F'A+,'"u="*G"A'N/*A'M"*(";0B&"
>K#
"
B/60A"+(M"(0A6+O"C0A"+"D+(M*M+B'"B0";'"G'O'DB'M"C0A"F+A*+(B"D+OO*(,R"
>K$
"
"
>K=
"
N."*/"4,7(*%$*9,5"19*,(91-*'(/*$155@'5,7(&"(-*,(91-8"3O+*A1"@*O'/@"*(@/B"D06@A*G'G"!-!##"*(B','AG"
>K>
"
s"$$"@0G*Y0(G"P*M'"P*B&"$="C'+B/A'G"+B"'+D&"@0G*Y0("\+("')+6@O'"*G",*F'("*("S)B'(M'M"?+B+"
>KK
"
_*,/A'"#+`R"^"M'B+*O'M"')@O+(+Y0("0C"'+D&"C'+B/A'"*G",*F'("*("1/@@O'6'(B+AH"h'B&0MG"/(M'A"
>KL
"
v?'GDA*@Y0("0C"@*O'/@"*(@/B"C'+B/A'GwR"3O+*A1"&+G"!Q"@*O'/@"C'+B/A'G"*("D0660("P*B&"3O+*A$-"+(M"
>KQ
"
!K"+MM*Y0(+O"C'+B/A'GR"7&'"!K"('P"C'+B/A'G"+A'"A'+M"D0/(BG"=LMQP-"=LMQ@-"=LBQP-"+(M"=LBQ@-"
>KV
"
P&'A'"="*G"'*B&'A"0C"B&'"(/DO'0YM'G"^-"3-"l-"+(M"7-"2hb"G/;GDA*@B"6'+(G"6+@@*(,"N/+O*BH"O0P'A"
>LW
"
B&+("#W"\hb{#W`-"2db"6'+(G";+G'"N/+O*BH"O0P'A"B&+("$W"\db{$W`-"+(M"c"+(M"5"6'+("B&'"
>L!
"
C0AP+AM"+(M"A'F'AG'"GBA+(M-"A'G@'DYF'OHR"7&'"A+Y0(+O'";'&*(M"B&'"('P"C'+B/A'G"*G"B&+B"*("3O+*A1-"
>L#
"
B&'"A'G/OBG"0C"@*O'/@5;+G'M"D+OO*(,"+(M"C/OO5+O*,(6'(B5;+G'M"D+OO*(,"+A'"BA/GB'M"'N/+OOH-"G0"B&'"
>L$
"
6+@@*(,"N/+O*BH"+(M";+G'"N/+O*BH"*(C0A6+Y0("B&+B"/G'M"B0";'"')DO/G*F'"B0"C/OO5+O*,(6'(B"D+OO*(,"
>L=
"
(''M"B0";'"+MM'M"B0"@*O'/@5;+G'M"D+OO*(,R"i/A"')@'A*6'(B"G&0P'M"B&+B"A'60F*(,"B&'"!K"('P"
>L>
"
C'+B/A'G"A'M/D'M"@A'D*G*0(";H"x#f"/G*(,">W)e#>)"0C"I33!$V>ed2R"3O+*A1"C/OO5+O*,(6'(B"*(@/B"
>LK
"
D06@A*G'G"$W-W$W"*(B','AG"s"G'F'("D&+(('OG-"'+D&"P*B&"$$"@0G*Y0(G"+(M"!$W"A0PG"B0"G/@@0AB"+B"
>LL
"
60GB"LK"B/60A"A'+MG-">#"(0A6+O"A'+MG-"+(M"#"'6@BH"A0PG"+G"G@+D'";'BP''("B/60A"+(M"(0A6+O"
>LQ
"
\+("')+6@O'"*G",*F'("*("S)B'(M'M"?+B+"_*,/A'"#;`R"2*8'"3O+*A$-"A+(M06"G/;G+6@O*(,"M0P("B0"B&'"
>LV
"
6+)*6/6"G/@@0AB'M"D0F'A+,'"*G"/G'M"+B"')D'GG*F'"D0F'A+,'GR"^"M'B+*O'M"')@O+(+Y0("0C"'+D&"
>QW
"
D&+(('O"*G",*F'("*("1/@@O'6'(B+AH"h'B&0MG"/(M'A"v?'GDA*@Y0("0C"C/OO5+O*,(6'(B"*(@/B"
>Q!
"
D&+(('OGwR"X(";0B&"*(@/BG-"B&'"D+(M*M+B'"F+A*+(B"*G"D'(B'A'M"+B"B&'"!Kth"@0G*Y0(R"q0G*Y0(G"
>Q#
"
/(D0F'A'M";H"+(H";+G'"*("C/OO5+O*,(6'(B"*(@/B"+A'"[OO'M"P*B&"k'A0R"
>Q$
"
"
>Q=
"
H"4,7(*%$*("1#'5*("-+%#C48"7&'"@*O'/@"+(M"C/OO5+O*,(6'(B"('BP0A8"+AD&*B'DB/A'"+(M"*6@0AB+(B"
>Q>
"
@+A+6'B'AG"+A'"G&0P("*("_*,/A'"$R"7&'"@*O'/@"('BP0A8"/G'G"BP0";*M*A'DY0(+O",+B'"A'D/AA'(B"/(*B"
>QK
"
\d*5l:E`"O+H'AG-"'+D&"P*B&"!#Q"+(M"!V#"/(*BGR"306@+A'M"B0"B&'"3O+*A$"@*O'/@"('BP0A8-"B&'"/G'"0C"
>QL
"
d*5l:E"*(GB'+M"0C";*M*A'DY0(+O"O0(,"G&0AB5B'A6"6'60AH"\d*5217h`"+AD&*B'DB/A'"A'M/D'M"
>QQ
"
BA+*(+;O'"@+A+6'B'AG"CA06"#->$#-VV>"B0"#-$WV->WL"+(M"6+BA*)"D06@/B+Y0(G"CA06"$R!!"B0"#R$Q"
>QV
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
;*OO*0(-";/B"*6@A0F'M"@'AC0A6+(D'"*("0/A"')@'A*6'(BR"7&'"C/OO5+O*,(6'(B"('BP0A8"*G"+"A'G*M/+O"
>VW
"
('/A+O"('BP0A8"\:'Ga'B`"D06@A*G*(,"B&A''"GB+(M+AM"A'G*M/+O";O0D8GR"^"D0(F0O/Y0(+O"O+H'A"*G"
>V!
"
+MM'M"*66'M*+B'OH";'C0A'"'+D&"A'G*M/+O";O0D8"B0"')@+(M"B&'"(/6;'A"0C"D&+(('OGR"X(";0B&"
>V#
"
('BP0A8G-"+"MA0@0/B"A+B'"+B"WR$"*G"G'B"C0A"B&'"p+g'('M"O+H'A"+(M"M'(G'"O+H'A"B0"@A'F'(B"
>V$
"
0F'A[j(,R"
>V=
"
"
>V>
"
>-"9*D8*>"'#).*$%#*'()"4-#'5*.'95%-;9"*4199%#-8"1B'@"$"*G"M'@*DB'M"*("_*,/A'"#MR"7&'"('/A+O"
>VK
"
('BP0A8G"')&*;*B'M",00M"@0P'A"*("M*GY(,/*G&*(,"A'+O"F+A*+(BG"CA06"C+OG'"@0G*YF'"D+(M*M+B'GR"
>VL
"
I0P'F'A-"/G'C/O"G*,(+OG"A'60B'"B0"+"F+A*+(B"D+(M*M+B'"+A'"(0B"D0F'A'M";H"B&'"D/AA'(B"('/A+O"
>VQ
"
('BP0A8"M'G*,(G"*("3O+*A1-"P&*D&"D0(G*M'AG"0(OH"B&'"p+(8*(,"!K;@"0C"+"D+(M*M+B'R"a0B+;OH-"'F'("
>VV
"
*C"B&'"p+(8*(,"P*(M0P"*G"')B'(M'M"B0">W;@-"*B"*G"GYOO"B00"G&0AB"C0A"+("+DD/A+B'"*(C'A'(D'"0C"P&*D&"
KWW
"
&+@O0BH@'"+"F+A*+(B"D+(M*M+B'";'O0(,G"B0"/G*(,"0(OH"B&'"('BP0A8G-";/B"B&'"('BP0A8G"P0/OM"
KW!
"
+OA'+MH";'"D06@/B+Y0(+OOH"*(C'+G*;O'"C0A"G06+YD"F+A*+(B"D+OO*(,R"X("3O+*A1-"@0GB5@A0D'GG*(,"GB'@"
KW#
"
$"*G"M'G*,('M"B0"A'M/D'"C+OG'"@0G*YF'"D+OO*(,"6*GB+8'G"6+M'";H"B&'"('BP0A8G";H"O'F'A+,*(,"
KW$
"
A'O+YF'OH"A'60B'",'A6O*('"F+A*+(BG"B0"[(M"B&'"D0AA'DB"+(D'GBA+O"&+@O0BH@'"C0A"+"G06+YD"F+A*+(BR"
KW=
"
^(H"G06+YD"F+A*+(B"D+OOG"B&+B"D+((0B";'"C0/(M"P*B&"+(D'GBA+O"&+@O0BH@'"G/@@0AB"+A'"GP*BD&'M"B0"
KW>
"
+("+AYC+DB"+(M"+A'"')DO/M'M"CA06"B&'"0/B@/BR"7&'"&+@O0B+,,'M"A'+MG"@A0M/D'M"*("GB'@"!"+A'"
KWK
"
/G'M"*("B&*G"GB'@R"_0A"+"G06+YD"F+A*+(B"B&+B"D0F'AG"+(H"&+@O0B+,,'M"A'+MG-"P'"A'N/*A'M"B&'"
KWL
"
G06+YD"F+A*+(B"B0"D0')*GB"P*B&"B&'"&'B'A0kH,0/G",'A6O*('"F+A*+(BG"O'GG"B&+("!WW";@"+P+H"0("*BG"
KWQ
"
O'Z"+(M"A*,&B"*("B&'"A'+MG"*("B&'"&+@O0BH@'",A0/@"B&'"G06+YD"F+A*+(B"G/@@0AY(,"A'+MG"P'A'"*(R"
KWV
"
^("')+6@O'"0C"+"C+OG'"@0G*YF'"G06+YD"F+A*+(B"[OB'A'M";H"B&*G"A/O'"*G",*F'("*("S)B'(M'M"?+B+"
K!W
"
_*,/A'"$+R"^"G06+YD"F+A*+(B"+B"D&A=T$Q-W!#-V=#"P+G"D+OO'M";H"B&'"BP0"('BP0A8GR"^"@&+G'M"
K!!
"
&'B'A0kH,0/G",'A6O*('"F+A*+(B"P+G"C0/(M"K!";@"O'Z"0C"B&'"G06+YD"D+OOR"7&A''"A'+MG"*("&+@O0BH@'"
K!#
"
#"B&+B"G/@@0AB'M"B&'"G06+YD"F+A*+(B"P'A'"C0/(M"(0B"B0"&+F'"B&'"&'B'A0kH,0/G",'A6O*('"F+A*+(BR"
K!$
"
7&/G-"B&'"G06+YD"F+A*+(B"P+G"D0(G*M'A'M"/(G/@@0AB'M";H"+("+(D'GBA+O"&+@O0BH@'R"_0A"+"G06+YD"
K!=
"
F+A*+(B"B&+B"D0F'AG"(0"&+@O0B+,,'M"A'+M-"*B"*G"@A0;+;OH";'D+/G'"B&'A'"+A'"(0",'A6O*('"F+A*+(BG"
K!>
"
0A"0(OH"&060kH,0/G"F+A*+(BG"*("B&'"F*D*(*BHR"X("B&*G"D+G'-"P'"A'N/*A'M"B&'"G06+YD"F+A*+(B"B0";'"
K!K
"
D0')*GY(,"P*B&"B&'"&060kH,0/G",'A6O*('"F+A*+(BG"O'GG"B&+("!WW";@"+P+H"0("*BG"O'Z"+(M"A*,&B"*("
K!L
"
+OO"G06+YD"F+A*+(B"G/@@0AY(,"A'+MGR"^("')+6@O'"0C"B&*G"*G",*F'("*("S)B'(M'M"?+B+"_*,/A'"$;R"^"
K!Q
"
G06+YD"F+A*+(B"+B"D&A!T!WW-K$#-!>Q"P+G"D+OO'M";H"B&'"BP0"('BP0A8GR"^"&060kH,0/G",'A6O*('"
K!V
"
F+A*+(B"P+G"C0/(M"$V";@"O'Z"0C"B&'"G06+YD"D+OOR"h/OY@O'"A'+MG"B&+B"G/@@0AB"B&'"G06+YD"F+A*+(B"
K#W
"
P'A'"C0/(M"(0B"B0"&+F'"B&'"&060kH,0/G",'A6O*('"F+A*+(BR"7&/G-"B&'"G06+YD"F+A*+(B"P+G"(0B"
K#!
"
D0(G*M'A'M"B0";'"G/@@0AB'M";H"+("+(D'GBA+O"&+@O0BH@'R"106+YD"F+A*+(BG"B&+B"M0"(0B"&+F'"+(H"
K##
"
,'A6O*('"F+A*+(BG"O'GG"B&+("!WW";@"+P+H"0("B&'*A"O'Z"0A"A*,&B"+A'"(0B"+@@O*D+;O'"*("B&*G"GB'@R"
K#$
"
"
K#=
"
J1-91-8"3O+*A1"G/@@0ABG"]3_"C0A6+B"0/B@/BR"106+YD"F+A*+(BG"+A'"6+A8'M"vq^11w"0A"v20Pb/+Ow"*C"
K#>
"
B&'"F+A*+(B"N/+O*BH"*G"O0P"\*R'R-"bE^2{Q-"D0([,/A+;O'";H"0@Y0(`-"0A"B&'H"+A'"[OB'A'M"*("GB'@"$R"_0A"
K#K
"
'+D&"F+A*+(B-"B&'"+OO'O'"CA'N/'(DH"+(M"G/@@0AY(,"D0F'A+,'"0C"B&'"A'C'A'(D'"+OO'O'"+(M"+OO"
K#L
"
+OB'A(+YF'"+OO'O'G"+A'"G&0P(R"7&'"0@Y0(G"v55@A*(By,'A6O*('yD+OOGw"+(M"v55@A*(ByA'CyD+OOGw"
K#Q
"
'(+;O'"0/B@/j(,",'A6O*('"F+A*+(BG"+(M"+AYC+DBG-"A'G@'DYF'OHR"
K#V
"
"
K$W
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
ONT library preparation and sequencing
K$!
"
l'(06*D"?a^"\,?a^`"0C"+"BA*@O'5(',+YF'";A'+GB"D+(D'A"\7ad3`"D'OO"O*('"\I33!$V>`"+(M"+"d"
K$#
"
OH6@&0DHB'5M'A*F'M"(0A6+O"D'OO"O*('"\I33!$V>d2`"CA06"B&'"G+6'"M0(0A"P'A'"@/AD&+G'M"CA06"
K$$
"
B&'"^6'A*D+("7H@'"3/OB/A'"30OO'DY0("\^733`R"l'(06*D"?a^"\,?a^`"0C"IlWW!"P+G"@/AD&+G'M"
K$=
"
CA06"B&'"30A*'OO"X(GYB/B'R"7&'"&*,&560O'D/O+A5P'*,&B",?a^"P+G"')+6*('M";H"a+(0MA0@-"b/;*B-"
K$>
"
+(M"WR$>f"+,+A0G'"'O'DBA0@&0A'G*G"C0A"*BG"D0(D'(BA+Y0(-"@/A*BH-"+(M"*(B',A*BHR"7&'",?a^"P+G"
K$K
"
B&'("CA+,6'(B'M"P*B&",7/;'"B0",'('A+B'"?a^"CA+,6'(BG"+@@A0)*6+B'OH"#W"8;"*("O'(,B&R"7&'G'"
K$L
"
CA+,6'(BG"P'A'"B&'(";'*(,"G'N/'(D'M"+B"BP0"G'N/'(D*(,"D'(B'AGT"IJE"+(M"a0F0,'('R"^B"IJE-"
K$Q
"
B&'"CA+,6'(BG"0C"I33!$V>-"I33!$V>d2-"+(M"IlWW!"P'A'"@A'@+A'M"+(M"O*,+B'M"P*B&"+"
K$V
"
G'N/'(D*(,"+M+@B'A"/G*(,"ia7„G"O*,+Y0("G'N/'(D*(,"8*B"]!="1bJ521J!!=R"7&'"O*,+B'M"G+6@O'G"
K=W
"
P'A'"G'N/'(D'M"0(":!WR=R!"qA06'B&Xia"p0PD'OOG"/G*(,"+"qA06'B&Xia"#"10O0"M'F*D'"+(M"
K=!
"
h*(Jai9"G0ZP+A'"F'AG*0("!R!QRW#-"C0A"VK"&R"^B"a0F0,'('-"B&'"CA+,6'(BG"0C"I33!$V>"+(M"
K=#
"
I33!$V>d2"P'A'"@A'@+A'M"+(M"O*,+B'M"P*B&"+"G'N/'(D*(,"+M+@B'A"/G*(,"ia7nG"O*,+Y0("
K=$
"
G'N/'(D*(,"8*B"]!#"1bJ521J!!#R"7&'"O*,+B'M"G+6@O'G"P'A'"G'N/'(D'M"0(":!WR="qA06'B&Xia"
K==
"
p0PD'OOG"/G*(,"qA06'B&Xia"=Q-"C0A"VK"&R"
K=>
"
"
K=K
"
Benchmarking
K=L
"
9'"/G'M"B&'"BA/B&"G'B"0C"G06+YD"F+A*+(BG"*("I33!$V>ed2",'('A+B'M"+(M"6+*(B+*('M";H"B&'"
K=Q
"
1Sb3#"D0(G0AY/6R"7&'"BA/B&"G'B"P+G"0AB&0,0(+OOH"F+O*M+B'M"P*B&"6/OY@O'"G'N/'(D*(,"A'@O*D+B'G"
K=V
"
CA06"6/OY@O'"G'N/'(D*(,"D'(B'AG"B&+B"D06@A*G'"0F'A"!->WW5C0OM"G'N/'(D*(,"M+B+"*("B0B+OR"9'"
K>W
"
/G'M"0(OH"B&'"G06+YD"F+A*+(BG"O+;'O'M"vI*,&30(Cw"\I*,&"30([M'(D'`"0A"vh'M30(Cw"\h'M*/6"
K>!
"
30([M'(D'`"+G"BA/B&R"106+YD"F+A*+(BG"O+;'O'M"v20P30(Cw"\20P"30([M'(D'-"]^_"zWRW>-"(0B"+"
K>#
"
@+AB"0C"B&'"BA/B&"G'B"+G"M'[('M";H"1Sb3#`"P'A'"(0B"/G'M"C0A";'(D&6+A8*(,R"X("B0B+O-"B&'A'"P'A'"
K>$
"
$V->KW"BA/B&"1a]G"+(M"!-V##"BA/B&"X(M'OGo"$V-==L"0C"B&'"1a]G"+(M"!-KW#"0C"B&'"X(M'OG"P'A'"
K>=
"
P*B&*("B&'"&*,&5D0([M'(D'"A',*0(G"M'[('M"*("+"dS?"[O'"@A0F*M'M";H"1Sb3#R"^"F+A*+(B"D+OO"P+G"
K>>
"
D0(G*M'A'M"D0AA'DB"0(OH"*C"*B"6+BD&'M";0B&"B&'",'(06'"@0G*Y0("+(M"F+A*+(B"+OO'O'"0C"B&'"BA/B&R"
K>K
"
_0A";0B&"B&'"ia7"+(M"XOO/6*(+";'(D&6+A8G-"G06'"BA/B&"F+A*+(BG"P'A'"')DO/M'M"C0A"B&'"C0OO0P*(,"
K>L
"
A'+G0(GR"_*AGB-"'F'("P*B&"B&'"&*,&"G'N/'(D*(,"D0F'A+,'-"G/D&"+G"L>RVL5C0OM"I33!$V>"P'"
K>Q
"
,'('A+B'M"C0A"B&'"ia7";'(D&6+A8G-"G06'"BA/B&"F+A*+(BG"GYOO"&+M"F'AH"O0P"0A"(0"D0F'A+,'-"0A"
K>V
"
&+M"(0"A'+M"G/@@0AY(,"B&'"F+A*+(B"+OO'O'R"7&'G'"BA/B&"F+A*+(BG"P0/OM"C+*O"+OO"B&'";'(D&6+A8G-"G0"
KKW
"
B&'H"G&0/OM";'"')DO/M'MR"1'D0(M-"G06'";'(D&6+A8G"B'GB'M"6/OY@O'"G'N/'(D*(,"D0F'A+,'G"+(M"
KK!
"
A'N/*A'M"G'N/'(D*(,"A'+M"G/;G+6@O*(,"CA06"B&'"C/OO"M+B+G'BR"7&'"G/;G+6@O*(,"@A0D'GG"6*,&B"
KK#
"
A'60F'"A'+MG"G/@@0AY(,"+"BA/B&"F+A*+(B"B0"+("')B'(B"B&+B"C'P"0A"(0"G/@@0AY(,"A'+MG"+A'"O'ZR"
KK$
"
7&*G"+m'DBG"'G@'D*+OOH"B&'"G06+YD"F+A*+(BG"B&+B"+OA'+MH"&+F'"+"O0P"]^_R"_0A"')+6@O'-"+"]^_"WRW>"
KK=
"
G06+YD"F+A*+(B"P*B&"#W5C0OM"D0F'A+,'"+(M"0('"A'+M"G/@@0AY(,"B&'"F+A*+(B"+OO'O'"D+(";'"A'M/D'M"
KK>
"
B0"]^_"W";H"A'60F*(,"t/GB"0('"A'+M"M/A*(,"G/;G+6@O*(,R"7&*G"A'M/D'G"B&'"N/+O*BH"0C"B&'"
KKK
"
;'(D&6+A8*(,"A'G/OBG-"'G@'D*+OOH"C0A"O0P"]^_"BA/B&"F+A*+(BG"P&'("G/;G+6@O'M"M+B+G'BG"+A'"/G'MR"
KKL
"
70"+OO'F*+B'"B&'"@A0;O'6-"+(H"BA/B&"F+A*+(BG"B&+B"&+F'"F'AH"O0P"]^_"\{WRW>`"0;G'AF'M"*("B&'"C/OO"
KKQ
"
M+B+G'B";'C0A'"G/;G+6@O*(,"G&0/OM";'"')DO/M'MR"1/66*(,"/@"B&'"BP0"A'+G0(G"+;0F'-"C0A"'+D&"
KKV
"
0C"B&'"C/OO"M+B+G'BG"P'"/G'M"*(";0B&"B&'"ia7"+(M"XOO/6*(+";'(D&6+A8G-"P'"')DO/M'M"BA/B&"
KLW
"
F+A*+(BG"B&+B"6+BD&'M"+(H"0C"B&'"C0OO0P*(,"DA*B'A*+"CA06";'(D&6+A8*(,T"!`"]^_"zWRW>-"#`"A'+MG"
KL!
"
G/@@0AY(,"B&'"F+A*+(B"+OO'O'"{$-"$`"B/60A"D0F'A+,'"{=-"+(M"=`"(0A6+O"D0F'A+,'"{=R"_0A"
KL#
"
GB+(M+AM*k+Y0(-"P'"/G'M"G06R@H-"@A0F*M'M"*("XOO/6*(+nG"I+@O0BH@'"306@+A*G0("700OG27"\F'AG*0("
KL$
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
FWR$R!#`"B0",'('A+B'"'F+O/+Y0("6'BA*DG-"*(DO/M*(,"_!51D0A'-"qA'D*G*0(-"+(M":'D+OO"+,+*(GB"B&'"
KL=
"
BA/B&"F+A*+(BGR"7&'"vD06@+A'yFDCw"G/;60M/O'"*("3O+*A1"@A0M/D'G"*M'(YD+O"A'G/OBG"B0"G06R@H-";/B"
KL>
"
+/B06+B'G"B&'"')DO/G*0("0C"/(N/+O*CH*(,"BA/B&"F+A*+(BGR"7&'"BA/B&"G'B"6+B'A*+OG"+A'"@/;O*DOH"
KLK
"
+F+*O+;O'"B0"B&'"D066/(*BHR"^OO"B00OG-"B&'*A"F'AG*0(-"+(M"D066+(M"O*('G"/G'M"+A'",*F'("*("B&'"
KLL
"
v3066+(M"O*('G"/G'Mw"G'DY0("*("1/@@O'6'(B+AH"a0B'GR"
KLQ
"
"
KLV
"
Computational performance
KQW
"
3O+*A1"P+G"PA*g'("*("qHB&0("+(M"3ccR"7&'"qHB&0("@+ABG"O'F'A+,'M"qHqH"C0A"G@''M"/@R"7&'"('/A+O"
KQ!
"
('BP0A8"*6@O'6'(B+Y0(G"/G'M"qH70AD&R"7A+*(*(,"3O+*A1"('/A+O"('BP0A8G"A'N/*A'G"+"&*,&5'(M"
KQ#
"
lqE-";/B"/G*(,"3O+*A1"C0A"G06+YD"F+A*+(B"D+OO*(,"A'N/*A'G"0(OH"+"3qER"_0A"B&'">W)e#>)"
KQ$
"
I33!$V>ed2"@+*A-"3O+*A1"[(*G&'M"A/((*(,"*("x>"&0/AG"C0A"ia7"M+B+"+(M"x#"&0/AG"C0A"XOO/6*(+"M+B+"
KQ=
"
\$Wf"GO0P'A"B&+("1BA'O8+#-";/B"C+GB'A"B&+("+OO"0B&'A"G&0AB5A'+M"G06+YD"F+A*+(B"D+OO'AG`-"/G*(,"
KQ>
"
BP0"!#5D0A'"X(B'O"…'0("1*OF'A"=!!K"@A0D'GG0AGR"7&'"6'60AH"C00B@A*(B"*G"O0P"+(M"*G"D0(BA0OO'M"+B"
KQK
"
O0P'A"B&+("!ld"@'A"3qER"_0A"60M'O"BA+*(*(,-"P'"B'GB'M"aF*M*+"l'_0AD'":7…"#WQW"7*-"$WVW-"+(M"
KQL
"
=WVW-"+(M"C0/(M"'+D&"('P"60M'O"@A0F*M'M"+"x$>f"G@''M"*(DA'+G'"CA06"B&'"@A'F*0/G"
KQQ
"
,'('A+Y0(R"
KQV
"
"
KVW
"
"
KV!
"
Code availability
KV#
"
3O+*A1"*G"0@'("G0/AD'"+(M"+F+*O+;O'"+B"&g@GTee,*B&/;RD06eIJE5d^2e3O+*A1"/(M'A"B&'"d1?"$5
KV$
"
3O+/G'"O*D'(G'R"7&'"A'G/OBG"*("B&*G"@+@'A"P'A'";+G'M"0("B&'"3O+*A1"*(*Y+O"A'O'+G'"\F'AG*0("WRWR!`R"
KV=
"
h/OY@O'"*(GB+OO+Y0("0@Y0(G"+A'"+F+*O+;O'"C0A"3O+*A1-"*(DO/M*(,"?0D8'A"+(M"1*(,/O+A*BHR"3O+*A1"&+G"
KV>
"
+OG0";''("*(DO/M'M"+G"B&'"G6+OO"F+A*+(B"D+OO'A"*("ia7nG"G06+YD"F+A*+(B"D+OO*(,"P0A8p0P29"G*(D'"
KVK
"
F'AG*0("WR!RWR"
KVL
"
"
KVQ
"
Data availability
KVV
"
7&'"O*(8G"B0"B&'"A'C'A'(D'",'(06'G-"BA/B&"G06+YD"F+A*+(BG-";'(D&6+A8*(,"6+B'A*+OG-"ia7-"+(M"
LWW
"
XOO/6*(+"M+B+"+A'",*F'("*("B&'"v?+B+"+F+*O+;*O*BHw"G'DY0("*("1/@@O'6'(B+AH"a0B'GR"^OO"+(+OHG*G"
LW!
"
0/B@/B-"*(DO/M*(,"B&'"]3_G"+(M"A/((*(,"O0,G-"*G"+F+*O+;O'"+B"
LW#
"
&g@TeePPPR;*0QRDGR&8/R&8eDO+*AGe+(+OHG*GyA'G/OBR"7&'"I33!$V>ed2"G'N/'(D*(,"M+B+",'('A+B'M"
LW$
"
*("B&*G"GB/MH"P+G"M'@0G*B'M"*("B&'"a3dX"G&0AB5A'+M"+AD&*F'"P*B&"+DD'GG*0("X?"q:.a^VQK#V#R"
LW=
"
"
LW>
"
Acknowledgements
LWK
"
:R2R"P+G"G/@@0AB'M";H"I0(,"J0(,":'G'+AD&"lA+(BG"30/(D*O",A+(BG"l:_"\!L!!$L#!`"+(M"7:1"\7#!5
LWL
"
LW>e#W5a`-"B&'"1&'(k&'("h/(*D*@+O"l0F'A(6'(B"l'('A+O"qA0,A+6"\.34.#W#!W$#=!$==W>W!>`-"
LWQ
"
B&'"E:3"C/(M"CA06"IJE-"+(M"i)C0AM"a+(0@0A'"7'D&(0O0,*'GR"
LWV
"
"
L!W
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
Author contribu.ons
L!!
"
:R2R"D0(D'*F'M"B&'"GB/MHR"%R%R"+(M":R2R"M'G*,('M"B&'"+O,0A*B&6G-"*6@O'6'(B'M"3O+*A1-"+(M"PA0B'"
L!#
"
B&'"@+@'AR"4R2R"+(M"7R59R2R"'F+O/+B'M"B&'";'(D&6+A8*(,"A'G/OBGR"^OO"+/B&0AG"A'F*G'M"B&'"
L!$
"
6+(/GDA*@BR"
L!=
"
'
L!>
"
Compe.ng interests
L!K
"
:R2R"A'D'*F'G"A'G'+AD&"C/(M*(,"CA06"ia7R"7&'"0B&'A"+/B&0AG"M'DO+A'"(0"D06@'Y(,"*(B'A'GBGR"
L!L
"
"
L!Q
"
" "
L!V
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
References
L#W
"
!R" 9'*(GB'*(-".RaR"'B"+OR"7&'"D+(D'A",'(06'"+BO+G"@+(5D+(D'A"+(+OHG*G"@A0t'DBR"='-1#"*
L#!
"
7"("3)4"IJ-"!!!$5!!#W"\#W!$`R"
L##
"
#R"q'A'A+5d'O-".R"'B"+OR"_A06"G06+YD"F+A*+(BG"B0P+AMG"@A'D*G*0("0(D0O0,HT"'F*M'(D'5MA*F'("
L#$
"
A'@0AY(,"0C"BA'+B6'(B"0@Y0(G"*("60O'D/O+A"B/60A";0+AMGR"F"(%&"*&"/,),(""BK-"!5!>"
L#=
"
\#W!Q`R"
L#>
"
$R" _+(,-"2R7R"'B"+OR"SGB+;O*G&*(,"D066/(*BH"A'C'A'(D'"G+6@O'G-"M+B+"+(M"D+OO"G'BG"C0A"
L#K
"
;'(D&6+A8*(,"D+(D'A"6/B+Y0("M'B'DY0("/G*(,"P&0O'5,'(06'"G'N/'(D*(,R"='-1#"*
L#L
"
2,%-").(%5%7;"HL-"!!>!5!!KW"\#W#!`R"
L#Q
"
=R" JA*G&(+6+D&+A*-"JR"'B"+OR"^DD/A+B'"G06+YD"F+A*+(B"M'B'DY0("/G*(,"P'+8OH"G/@'AF*G'M"
L#V
"
M''@"O'+A(*(,R"='-1#"*G%&&1(,)'3%(4"BH-"=#=Q"\#W##`R"
L$W
"
>R"1+&A+'*+(-"1RhRSR"'B"+OR"?''@"D0(F0O/Y0(+O"('/A+O"('BP0A8G"C0A"+DD/A+B'"G06+YD"
L$!
"
6/B+Y0("M'B'DY0(R"='-1#"*)%&&1(,)'3%(4"BK-"!W=!"\#W!V`R"
L$#
"
KR" a+Ak*G*-"lR"'B"+OR"l'(06'5P*M'"G06+YD"F+A*+(B"D+OO*(,"/G*(,"O0D+O*k'M"D0O0A'M"M'"dA/*t("
L$$
"
,A+@&GR"G%&&1(,)'3%(4*2,%5%7;"B-"#W"\#W!Q`R"
L$=
"
LR" _+(-"4R"'B"+OR"h/1ST"+DD0/(Y(,"C0A"B/60A"&'B'A0,'('*BH"/G*(,"+"G+6@O'5G@'D*[D"'AA0A"
L$>
"
60M'O"*6@A0F'G"G'(G*YF*BH"+(M"G@'D*[D*BH"*("6/B+Y0("D+OO*(,"CA06"G'N/'(D*(,"M+B+R"
L$K
"
F"(%&"*2,%5%7;"BM-"!5!!"\#W!K`R"
L$L
"
QR" 3*;/OG8*G-"JR"'B"+OR"1'(G*YF'"M'B'DY0("0C"G06+YD"@0*(B"6/B+Y0(G"*("*6@/A'"+(M"
L$Q
"
&'B'A0,'('0/G"D+(D'A"G+6@O'GR"='-1#"*2,%-").(%5%7;"HB-"#!$5#!V"\#W!$`R"
L$V
"
VR" 2+AG0(-"?RSR"'B"+OR"106+YD1(*@'AT"*M'(Y[D+Y0("0C"G06+YD"@0*(B"6/B+Y0(G"*("P&0O'"
L=W
"
,'(06'"G'N/'(D*(,"M+B+R"Q,%,($%#&'3)4"EN-"$!!5$!L"\#W!#`R"
L=!
"
!WR" J*6-"1R"'B"+OR"1BA'O8+#T"C+GB"+(M"+DD/A+B'"D+OO*(,"0C",'A6O*('"+(M"G06+YD"F+A*+(BGR"='-1#"*
L=#
"
&"-.%/4"BJ-">V!5>V="\#W!Q`R"
L=$
"
!!R" _A''M-"?R-"q+(-":R"†"^OM+(+-":R"7aGD0@'T"+DD/A+B'"M'B'DY0("0C"G06+YD"6/B+Y0(G"P*B&"
L==
"
&+@O0BH@'5;+G'M"F+A*+(B"D+(M*M+B'"M'B'DY0("+(M"6+D&*('"O'+A(*(,"[OB'A*(,R"2,%#R,6-"
L=>
"
#>WK=L"\#W!Q`R"
L=K
"
!#R" 3008'-"?RqR-"9'M,'-"?R3R"†"2/(B'A-"lR"^"/(*['M"&+@O0BH@'5;+G'M"6'B&0M"C0A"+DD/A+B'"
L=L
"
+(M"D06@A'&'(G*F'"F+A*+(B"D+OO*(,R"='-1#"*2,%-").(%5%7;"HL-"QQ>5QV#"\#W#!`R"
L=Q
"
!$R" 2+*-"%R"'B"+OR"]+A?*DBT"+"(0F'O"+(M"F'AG+YO'"F+A*+(B"D+OO'A"C0A"(')B5,'('A+Y0("G'N/'(D*(,"*("
L=V
"
D+(D'A"A'G'+AD&R"=1)5",)*'),/4*#"4"'#)."II-"'!WQ5'!WQ"\#W!K`R"
L>W
"
!=R" J0F+8+-"1R-"i/-"1R-".'(*8'-"JRhR"†"1D&+Bk-"hR3R"^@@A0+D&*(,"D06@O'B'",'(06'G-"
L>!
"
BA+(GDA*@B06'G"+(M"'@*506'G"P*B&"+DD/A+B'"O0(,5A'+M"G'N/'(D*(,R"='-1#"*S"-.%/4"EK-"
L>#
"
!#5!K"\#W#$`R"
L>$
"
!>R"^6'/A-"^R-"JO00GB'A6+(-"9RqR"†"I'GB+(M-"hR1R"1*(,O'560O'D/O'"G'N/'(D*(,T"B0P+AMG"
L>=
"
DO*(*D+O"+@@O*D+Y0(GR"N#"(/4*,(*2,%-").(%5%7;"HM-"L#5Q>"\#W!V`R"
L>>
"
!KR" a+(0@0A'"b#Wc"D&'6*GBAH-"&g@GTee(+(0@0A'B'D&RD06eN#W@O/G5D&'6*GBAHR""\#W!V`R"
L>K
"
!LR"_0)-"SR.R-":'*M5d+HO*GG-"JR1R-"S60(M-"hR.R"†"20';-"2R^R"^DD/A+DH"0C"(')B",'('A+Y0("
L>L
"
G'N/'(D*(,"@O+}0A6GR"="R-*7"("#'3%(L*4"T1"(),(7*U*'995,)'3%(4"B"\#W!=`R"
L>Q
"
!QR" 2/0-":R-"1'MO+k'D8-"_R.R-"2+6-"7R59R"†"1D&+Bk-"hR3R"^"6/OY5B+G8"D0(F0O/Y0(+O"M''@"('/A+O"
L>V
"
('BP0A8"C0A"F+A*+(B"D+OO*(,"*("G*(,O'"60O'D/O'"G'N/'(D*(,R"='-1#"*)%&&1(,)'3%(4"BK-"
LKW
"
VVQ"\#W!V`R"
LK!
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
!VR" q0@O*(-":R"'B"+OR"^"/(*F'AG+O"1aq"+(M"G6+OO5*(M'O"F+A*+(B"D+OO'A"/G*(,"M''@"('/A+O"
LK#
"
('BP0A8GR"='-1#"*2,%-").(%5%7;"HO-"VQ$5VQL"\#W!Q`R"
LK$
"
#WR" 9+,('A-".R"'B"+OR"d'(D&6+A8*(,"D&+OO'(,*(,"G6+OO"F+A*+(BG"P*B&"O*(8'M"+(M"O0(,"A'+MGR"G"55*
LK=
"
F"(%&,)4"E-"!WW!#Q"\#W##`R"
LK>
"
#!R" 2/0-":R"'B"+OR"S)@O0A*(,"B&'"O*6*B"0C"/G*(,"+"M''@"('/A+O"('BP0A8"0("@*O'/@"M+B+"C0A"
LKK
"
,'A6O*('"F+A*+(B"D+OO*(,R"='-1#"*S').,("*?(-"55,7"()""E-"##W5##L"\#W#W`R"
LKL
"
##R" %&'(,-"%R"'B"+OR"1H6@&0(*k*(,"@*O'/@"+(M"C/OO5+O*,(6'(B"C0A"M''@"O'+A(*(,5;+G'M"O0(,5
LKQ
"
A'+M"F+A*+(B"D+OO*(,R"='-1#"*G%&91-'3%('5*>),"()""E-"LVL5QW$"\#W##`R"
LKV
"
#$R" 1&+[(-"JR"'B"+OR"I+@O0BH@'5+P+A'"F+A*+(B"D+OO*(,"P*B&"qSqqS:5h+A,*(5?''@]+A*+(B"
LLW
"
'(+;O'G"&*,&"+DD/A+DH"*("(+(0@0A'"O0(,5A'+MGR"='-1#"*&"-.%/4"BN-"!$##5!$$#"\#W#!`R"
LL!
"
#=R" 160O8+-"hR"'B"+OR"306@A'&'(G*F'"GBA/DB/A+O"F+A*+(B"M'B'DY0(T"CA06"60G+*D"B0"
LL#
"
@0@/O+Y0(5O'F'OR"Q,%VR,6-"#W##R#WW=R"#WW=R=QLW>>"\#W##`R"
LL$
"
#>R" 1&*A+*G&*-"4R"'B"+OR"qA'D*G'"D&+A+DB'A*k+Y0("0C"G06+YD"D06@O')"GBA/DB/A+O"F+A*+Y0(G"CA06"
LL=
"
B/60AeD0(BA0O"@+*A'M"O0(,5A'+M"G'N/'(D*(,"M+B+"P*B&"(+(060(GFR"=1)5",)*:),/4*
LL>
"
V"4"'#).-",8+M>#K"\#W#$`R"
LLK
"
#KR" SP*(,-"^R?R"'B"+OR"306;*(*(,"B/60A",'(06'"G*6/O+Y0("P*B&"DA0PMG0/AD*(,"B0"
LLL
"
;'(D&6+A8"G06+YD"G*(,O'5(/DO'0YM'5F+A*+(B"M'B'DY0(R"='-1#"*&"-.%/4"BE-"K#$5K$W"
LLQ
"
\#W!>`R"
LLV
"
#LR" JA/GD&'-"qR"'B"+OR"d'GB"@A+DYD'G"C0A";'(D&6+A8*(,",'A6O*('"G6+OO5F+A*+(B"D+OOG"*("&/6+("
LQW
"
,'(06'GR"='-1#"*2,%-").(%5%7;"HM-">>>5>KW"\#W!V`R"
LQ!
"
#QR" 1&*A+*G&*-"4R"'B"+OR"qA'D*G'"D&+A+DB'A*k+Y0("0C"G06+YD"D06@O')"GBA/DB/A+O"F+A*+Y0(G"CA06"
LQ#
"
@+*A'M"O0(,5A'+M"G'N/'(D*(,"M+B+"P*B&"(+(060(GFR"Q,%VR,6-"#W#WR#WWLR"#W##R#!=#K#"
LQ$
"
\#W#W`R"
LQ=
"
#VR" a+(0@0A'"SqX#hS"2+;G-"&g@GTee,*B&/;RD06e'@*#6'5O+;GePC5G06+YD5F+A*+Y0(R""\#W#$`R"
LQ>
"
$WR" 1+&A+'*+(-"1RhRSR"'B"+OR"^D&*'F*(,"A0;/GB"G06+YD"6/B+Y0("M'B'DY0("P*B&"M''@"O'+A(*(,"
LQK
"
60M'OG"M'A*F'M"CA06"A'C'A'(D'"M+B+"G'BG"0C"+"D+(D'A"G+6@O'R"F"(%&"*Q,%5%7;"EH-"!#"
LQL
"
\#W##`R"
LQQ
"
$!R" 7+A+;*D&*-"hR"'B"+OR"^"@A+DYD+O",/*M'"B0"D+(D'A"G/;DO0(+O"A'D0(GBA/DY0("CA06"?a^"
LQV
"
G'N/'(D*(,R"='-1#"*&"-.%/4"BN-"!==5!>>"\#W#!`R"
LVW
"
$#R"2*(-".R5IR-"3&'(-"2R53R-"4/-"1R53R"†"I/+(,-"4R57R"20(,q&+G'T"+("/OBA+5C+GB"D&A060G06'5GD+O'"
LV!
"
@&+G*(,"+O,0A*B&6"C0A"G6+OO"+(M"O+A,'"F+A*+(BGR"Q,%,($%#&'3)4"HN-"!Q!K5!Q##"\#W##`R"
LV#
"
$$R" q+g'AG0(-"hR"'B"+OR"9&+BGI+@T"P'*,&B'M"&+@O0BH@'"+GG'6;OH"C0A"C/B/A'5,'('A+Y0("
LV$
"
G'N/'(D*(,"A'+MGR"W%1#('5*%$*G%&91-'3%('5*Q,%5%7;"EE-"=VQ5>WV"\#W!>`R"
LV=
"
"
LV>
"
" "
LVK
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
Figures
LVL
"
"
LVQ
"
\+`"
LVV
"
"
QWW
"
"
QW!
"
\;`"
QW#
"
"
QW$
"
"
QW=
"
\D`"
QW>
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
"
QWK
"
"
QWL
"
Figure 1. Overview of ClairS training data synthesis workflow.
QWQ
"
\+`"7&'"P0A8p0P"M'60(GBA+B'G"&0P"B0"@A0M/D'"GH(B&'YD"G06+YD"F+A*+(BG"/G*(,"BP0";*0O0,*D+OOH"
QWV
"
/(A'O+B'M"G+6@O'G"P*B&"8(0P("BA/B&",'A6O*('"F+A*+(BG"C0A"3O+*A1"60M'O"BA+*(*(,R"X("B&*G"GB/MH-"
Q!W
"
G@'D*[D+OOH-"P'"/G'M"QW)"ia7"9l1"M+B+"0C"lX^d"IlWW#"+G"G+6@O'"^-"+(M">W)"IlWW!"+G"G+6@O'"
Q!!
"
dR"_*AGB-",'A6O*('"F+A*+(BG"FA"+(M"FB"P'A'"M'[('M"+G"8(0P("BA/B&",'A6O*('"F+A*+(BG"*("G+6@O'"^"
Q!#
"
+(M"d",*F'(";H"lX^dR"FA"+(M"FB"*(DO/M'";0B&"&060kH,0/G"+(M"&'B'A0kH,0/G",'A6O*('"F+A*+(BG"0C"
Q!$
"
+"G+6@O'R"70",'('A+B'"GH(B&'YD"B/60A"F+A*+(BG"N"+(M"GH(B&'YD"(0A6+O"F+A*+(BG"=Ae=B"C0A"'+D&"
Q!=
"
G+6@O'-"B&'"+O*,(6'(BG"P'A'"G@O*B"*(B0"G6+OO'A"D&/(8G"P*B&"=)"D0F'A+,'"'+D&R"7&'(-"B&'"D&/(8G"
Q!>
"
CA06";0B&"G+6@O'G"P'A'"D06;*('M"+(M"B&'"F+A*+(BG"D+OO'M"CA06"B&'6"P'A'"M'[('M"+G"NR"9*B&"
Q!K
"
B&'"p')*;*O*BH"0C"D06;*(*(,"+(H"(/6;'A"0C"D&/(8G"CA06";0B&"G+6@O'G-"N"'m'DYF'OH"D0F'A'M"
Q!L
"
F+A*+(BG"D+OO'M"+B"M*m'A'(B"D0F'A+,'G"+(M"]^_R"1*6*O+AOH-"B&'"D&/(8G"CA06"+"G+6@O'"P'A'"
Q!Q
"
D06;*('M"+B"6/OY@O'"D0F'A+,'G"C0A"D+OO*(,"GH(B&'YD"(0A6+O"F+A*+(BG"=A"+(M"=BR"9*B&"+"G6+OO"
Q!V
"
(/6;'A"0C"D&/(8G"CA06"+(0B&'A"G+6@O'"D06;*('M"*(B0"+"GH(B&'YD"(0A6+O-"=A"+(M"=B"'m'DYF'OH"
Q#W
"
D0F'A'M"M*m'A'(B"D0(B+6*(+Y0("O'F'OGR"7&'"F+A*+(BG"FA-"FB-"N-"=A"+(M"=B"P'A'"B&'("/G'M"B0"
Q#!
"
,'('A+B'"C0/A"D+B',0A*'G"0C"F+A*+(BG"s"v106+YDw-"vl'A6O*('-"v^AYC+DBw-"+(M"va0A6+O50(OHw"s"
Q##
"
P*B&"M*m'A'(B"A/O'GR"106+YD-"l'A6O*('-"+(M"^AYC+DB"6+BD&"B&'"B&A''"D+B',0A*'G"*("B&'"*(C'A'(D'"
Q#$
"
B+G8"0C"B&'"3O+*A1"('BP0A8"+AD&*B'DB/A'R"7&'"F+A*+(BG"0C"B&'"B&A''"D+B',0A*'G"P'A'"/G'M"C0A"
Q#=
"
60M'O"BA+*(*(,R"9&'("/G*(,"G+6@O'"d"+G"B/60A"+(M"^"+G"(0A6+O-"106+YD"*G"M'[('M"+G"v\N5=A`"
L
"
Q#>
"
\FB5FA`w-"*R'R-"F+A*+(BG"B&+B"P'A'"!`"C0/(M"*("GH(B&'YD"B/60A"No"#`"(0B"C0/(M"*("GH(B&'YD"(0A6+O"
Q#K
"
=Ao"$`"C0/(M"+G"+",'A6O*('"F+A*+(B"*("FBo"0A"=`"(0B"C0/(M"*("FAR"l'A6O*('"*G"M'[('M"+G"vN*
L
"=A*
L
"
Q#L
"
FA*
L
*FBw-"*R'R-"F+A*+(BG"B&+B"P'A'"C0/(M"*("+OO"N-"=A-"FA-"+(M"FBR"^AYC+DB"*G"M'[('M"+G"vN5=A5FA5
Q#Q
"
FBw-"P&*D&"G*,(*['G"B&'"F+A*+(BG"C0/(M"0(OH"*("N"+(M"(0B"*("B&'",'A6O*('G"0A"GH(B&'YD"(0A6+OR"
Q#V
"
9&'("/G*(,"G+6@O'"^"+G"B/60A"+(M"d"+G"(0A6+O-"B&'"M'[(*Y0(G"A'6+*("B&'"G+6'"')D'@B"C0A"
Q$W
"
GP*BD&*(,"B&'"G/;GDA*@BGR""
Q$!
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
\;`"7&'"]^_"M*GBA*;/Y0("0C"B&'"GH(B&'YD"G06+YD"1a]G"+B"B&A''"M*m'A'(B"G*6/O+B'M"B/60A"
Q$#
"
@/A*Y'G"\!WWf-">Wf-"+(M"#>f`-"/G*(,"'*B&'A"IlWW!eIlWW#"0A"IlWW#eIlWW!"+G"B/60Ae(0A6+OR"
Q$$
"
1*(D'";0B&"&'B'A0kH,0/G"+(M"&060kH,0/G"F+A*+(BG"P'A'"/G'M"*("GH(B&'G*G-"+B"!WWf"B/60A"
Q$=
"
@/A*BH-"B&'"F+A*+(BG"P'A'",+B&'A'M"+B"WR>"+(M"!RW"]^_R"7&'"M*GBA*;/Y0("G&0P'M",00M"D0F'A+,'"0C"
Q$>
"
BH@*D+O"G06+YD"1a]"]^_";H"B&'"GH(B&'YD"1a]GR""
Q$K
"
"
Q$L
"
\D`"7&'";A'+8M0P("0C"B&'"(/6;'A"0C"GH(B&'YD"F+A*+(BG"C0A"BA+*(*(,R"7&'"(/6;'AG"!`"/G*(,"'*B&'A"
Q$Q
"
IlWW#eIlWW!"0A"IlWW!eIlWW#"+G"B/60Ae(0A6+O-"+(M"#`"0C"B&'"B&A''"D+B',0A*'G"106+YD-"
Q$V
"
l'A6O*('-"+(M"^AYC+DB-"+G"M'[('M"*("G/;[,/A'"+-"+A'"G&0P(R"7&'"(/6;'A"0C"106+YD"D+B',0A*'G"
Q=W
"
*G"C/AB&'A"M*F*M'M"*(B0"B&0G'"GH(B&'G*k'M"CA06"'*B&'A"&060kH,0/G"1aqG"0A"&'B'A0kH,0/G"1aqGR"
Q=!
"
7&'G'"(/6;'AG"')@O+*("P&H"*(DO/M*(,"&'B'A0kH,0/G"1aqG"*("B&'"GH(B&'G*G"*G"'GG'(Y+O"B0"'(G/A'"+"
Q=#
"
G/|D*'(B"(/6;'A"0C"GH(B&'YD"G06+YD"F+A*+(BG"C0A"60M'O"BA+*(*(,R""
Q=$
"
"
Q==
"
"
Q=>
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
"
Q=K
"
"
Q=L
"
Figure 2. Overview of the ClairS somatic variant calling workflow.
Q=Q
"
\+`"7&'"P0A8p0P"*OO/GBA+B'G"B&'"B&A''"GB'@G"0C"3O+*A1R"X("GB'@"!-"3O+*A1"/G'G"3O+*A$"+(M"20(,q&+G'"
Q=V
"
C0A",'A6O*('"F+A*+(B"D+OO*(,-"@&+G*(,"+(M"A'+M"&+@O0B+,,*(,R"7&'"@A0D'GG'M"+O*,(6'(BG"+A'"B&'("
Q>W
"
/G'M"C0A";0B&"@*O'/@5"+(M"C/OO5+O*,(6'(B5;+G'M"G06+YD"F+A*+(B"D+OO*(,"*("GB'@"#R"1B'@"$"*(F0OF'G"
Q>!
"
@0GB5@A0D'GG*(,"[OB'AG"B&+B"'O*6*(+B'"G06+YD"F+A*+(B"D+OO*(,G"*C"+("+(D'GBA+O"&+@O0BH@'"\6+B'A(+O"
Q>#
"
0A"@+B'A(+O`"CA06"P&*D&"B&'"G06+YD"F+A*+(B"D0/OM"0A*,*(+B'"D+((0B";'"C0/(MR"7&'"M'B+*OG"0C"
Q>$
"
GB'@G"!-"#-"+(M"$"+A'"G&0P("*("G/;[,/A'G";-"D-"+(M"MR"\;`"1B'@"!"M'B+*OGR"3O+*A$"*G"+@@O*'M"B0";0B&"
Q>=
"
B/60A"+(M"(0A6+O"G+6@O'G"C0A",'A6O*('"F+A*+(B"D+OO*(,R"I*,&5N/+O*BH"&'B'A0kH,0/G",'A6O*('"
Q>>
"
F+A*+(BG"G&+A'M";H";0B&"G+6@O'G"+A'"G'O'DB'M"+(M"/G'M";H"20(,q&+G'"B0"@&+G'"B&'",'A6O*('"
Q>K
"
F+A*+(BG"C0/(M"*("B&'"B/60A"G+6@O'R"EG*(,"B&'"@&+G'M",'A6O*('"F+A*+(BG-"B&'"B/60A"A'+MG"+A'"
Q>L
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
B&'("&+@O0B+,,'M"B0";'O0(,"B0"'*B&'A"&+@O0BH@'"!-"#-"0A"/(8(0P(R"\D`"1B'@"#"M'B+*OGR"7&'"
Q>Q
"
@A0D'GG'M"+O*,(6'(BG"CA06"GB'@"!"+A'"C'M"*(B0";0B&"B&'"@*O'/@5;+G'M"F+A*+(B5D+OO*(,"('/A+O"
Q>V
"
('BP0A8"+(M"B&'"C/OO5+O*,(6'(B";+G'M"F+A*+(B5D+OO*(,"('/A+O"('BP0A8R"i("+"G*(,O'"G06+YD"F+A*+(B"
QKW
"
D+(M*M+B'-";0B&"('BP0A8G",*F'"A'G@'DYF'"@A'M*DY0(G"0("B&'"@A0;+;*O*BH"0C"B&A''"D+B',0A*'GT"
QK!
"
v106+YDw-"vl'A6O*('w-"+(M"v^AYC+DBwR"7&'"@A'M*DY0(G"+A'"B&'("6'A,'M"+DD0AM*(,"B0"+"G'B"0C"
QK#
"
A/O'G"*(BA0M/D'M"*("B&'"h'B&0M"G'DY0(R"\M`"1B'@"$"M'B+*OGR"7&'"G06+YD"F+A*+(BG"D+OO'M"*("GB'@"#"
QK$
"
+A'"')+6*('M"B0"M'B'A6*('"*C"B&'H"+A'"G/@@0AB'M";H"+("+(D'GBA+O"&+@O0BH@'R"^(D'GBA+O"
QK=
"
&+@O0BH@'G-"P&*D&"D+(";'"'*B&'A"6+B'A(+O"0A"@+B'A(+O-"+A'"M'A*F'M"/G*(,",'A6O*('"F+A*+(BGR"^"
QK>
"
G06+YD"F+A*+(B"*G"D0(G*M'A'M"G/@@0AB'M";H"+("+(D'GBA+O"&+@O0BH@'"*C"B&'"&+@O0BH@'"D0(B+*(*(,"
QKK
"
B&'"G06+YD"F+A*+(B"*G";'O*'F'M"B0"0A*,*(+B'"CA06"0('"0C"B&'"+(D'GBA+O"&+@O0BH@'GR"
QKL
"
"
QKQ
"
"
QKV
"
"
QLW
"
"
QL!
"
Figure 3. The ClairS neural network architecture.
QL#
"
d0B&"\+`"B&'"@*O'/@"('BP0A8"+(M"\;`"C/OO5+O*,(6'(B"('BP0A8"/G'"+O*,(6'(BG"0C";0B&"B&'"B/60A"
QL$
"
+(M"(0A6+O"G+6@O'G"+G"*(@/BR"7'(G0AG"+A'"DA'+B'M"CA06";0B&"G+6@O'G"/G*(,"6'B&0MG"M'B+*O'M"*("
QL=
"
B&'"h'B&0M"G'DY0(-"+(M"+A'"B&'("D0(D+B'(+B'MR"7&'"B'(G0AG"+A'"B&'("@A0D'GG'M";H"B&'*A"
QL>
"
A'G@'DYF'"('/A+O"('BP0A8"C0A"*(C'A'(D'R"d0B&"('BP0A8G"0/B@/B"B&'"@A0;+;*O*BH"0C"B&A''"
QLK
"
D+B',0A*'GT"vl'A6O*('w-"v106+YDw-"+(M"v^AYC+DBwR"7&'"G'N/'(D'"0C"O+H'AG"+(M"O+H'A"
QLL
"
D0([,/A+Y0(G"+A'"G&0P(R"7&'"O'g'AG"D-"G-"+(M"8-"A'@A'G'(B"D&+(('O-"GBA*M'-"+(M"8'A('O-"
QLQ
"
A'G@'DYF'OHR"
QLV
"
"
QQW
"
+"
QQ!
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
"
QQ#
"
"
QQ$
"
;"
QQ=
"
"
QQ>
"
"
QQK
"
"
QQL
"
D"
QQQ
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
"
QQV
"
"
QVW
"
Figure 4. ONT HCC1395/BL dataset benchmarking results.
QV!
"
\+`"7&'"@A'D*G*0(5A'D+OO"D/AF'"0C"M*m'A'(B"D06;*(+Y0(G"0C"B/60A"+(M"(0A6+O"D0F'A+,'R"7&'"M0B"
QV#
"
0("'+D&"M+G&'M"O*('"G&0PG"P&'A'"B&'";'GB"_!5GD0A'"P+G"+D&*'F'MR"\;`"7&'"@'AC0A6+(D'"0C"3O+*A1"
QV$
"
+B"6/OY@O'"]^_"A+(,'G";'(D&6+A8'M"0("B&'"ia7"I33!$V>ed2"M+B+G'BR"X("B&'"[AGB"A0P-"#>-">W)-"
QV=
"
+(M"L>)"B/60A"P'A'"B'GB'M-"P*B&"B&'"(0A6+O"D0F'A+,'"[)'M"+B"#>)R"X("B&'"G'D0(M"A0P-"#W)-"#>)-"
QV>
"
+(M"$W)"0C"(0A6+O"P'A'"B'GB'M-"P*B&"B/60A"D0F'A+,'"[)'M"+B">W)R"]+A*+(B"N/+O*BH"D/B0m"Q"
QVK
"
\@A*0A*Yk'5A'D+OO"60M'`"P+G"/G'MR"\D`"7&'"@A'D*G*0(5A'D+OO"D/AF'"0C"M*m'A'(B"B/60Ae(0A6+O"@/A*BH"
QVL
"
D06;*(+Y0(G"P*B&"B/60A"D0F'A+,'"[)'M"+B">W)"+(M"(0A6+O"D0F'A+,'"[)'M"+B"#>)R"
QVQ
"
"
QVV
"
"
VWW
"
"
VW!
"
"
VW#
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
Figure 5. Categorizing the FPs and FNs in ClairS
VW$
"
7&'"@*'"D&+ABG"G&0P"B&'"M*GBA*;/Y0("0C"A'+G0(G"C0A"B&'"_qG"+(M"_aG"*("3O+*A1R"^">W)e#>)"ia7"
VW=
"
I33!$V>ed2"M+B+G'B"P+G"/G'M"C0A"D+OO*(,R"$WW"_qG"+(M"$WW"_aG"P'A'"A+(M06OH"D&0G'("CA06"B&'"
VW>
"
A'G/OBG"+(M"+(+OHk'MR"7&'"B+(M'6"A'@'+B-"O0P"D06@O')*BH-"&060@0OH6'A-"+(M"G',6'(B+O"
VWK
"
M/@O*D+Y0("A',*0(G"P'A'"M'[('M"/G*(,"lX^d"F$RW"l'(06'"1BA+Y[D+Y0(R"^60(,"B&'"D+B',0A*'G-"
VWL
"
v')D'GG*F'"6*G6+BD&'G"*("+O*,(6'(Bw"+(M"v*(G/|D*'(B"(0A6+O"D0F'A+,'w"P'A'"M'D*M'M"
VWQ
"
6+(/+OOH-"*R'R-"P*B&0/B"D'AB+*("D/B50mGR"vS)D'GG*F'"6*G6+BD&'G"*("+O*,(6'(Bw"P+G",*F'("*C"+("'H'"
VWV
"
D&'D8"0C"B&'"+O*,(6'(BG"A'F'+O'M"')D'GG*F'"*(D0(G*GB'(B"6*G6+BD&'G"B&+("/G/+O"+O*,(6'(BG"P*B&"
V!W
"
+"BA/'"G06+YD"F+A*+(BR"vX(G/|D*'(B"(0A6+O"D0F'A+,'w"P+G",*F'("P&'("+",'A6O*('"F+A*+(B"G*,(+O"
V!!
"
')*GB'M"*(";0B&"B/60A"+(M"(0A6+O-";/B"B&'"D0F'A+,'"0C"(0A6+O"P+G"O0P-"G0"B&'",'A6O*('"F+A*+(B"
V!#
"
G*,(+O"*("(0A6+O"P+G"0;F*0/GOH"P'+8'A"B&+("*("B/60AR"
V!$
"
"
V!=
"
"
V!>
"
+"
V!K
"
"
V!L
"
"
V!Q
"
;"
V!V
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
"
V#W
"
"
V#!
"
D"
V##
"
"
V#$
"
"
V#=
"
M"
V#>
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
"
V#K
"
"
V#L
"
Figure 6. Ilumina HCC1395/BL dataset benchmarking results.
V#Q
"
\+`"7&'"@A'D*G*0(5A'D+OO"D/AF'"0C"I33!$V>ed2"G&0AB5A'+M"M+B+G'BG"CA06"G*)"1Sb3#"G0/AD'G"\a1T"
V#V
"
a0F+1'N"+B"XOO/6*(+-"a3T"I*1'N"+B"a+Y0(+O"3+(D'A"X(GYB/B'-"X2T"I*1'N"+B"XO/6*(+-"S^T"I*1'N"+B"
V$W
"
S/A0@'+("X(CA+GBA/DB/A'"C0A"7A+(GO+Y0(+O"h'M*D*('-"_?T"I*1'N"+B"_/M+("E(*F'AG*BH-"a]T"I*1'N"+B"
V$!
"
a0F+AYG`"/G*(,"'*,&B"B00OG"\1BA'O8+#-"2+(D'B-"h/B'DB#-"a'/G06+YD-"iDB0@/G-"106+YD1(*@'A-"
V$#
"
]+A('B-"3O+*A1`R"]+A*+(BG"P'A'"A+(8'M";H"1BA'O8+#"s"106+YDS]1-"h/B'DB#"s"72i?-"]+Aa'B"s"1D0A'-"
V$$
"
106+YD1(*@'A"s"113-"+(M"0B&'A"D+OO'AG"s"bE^2R"7&'"M0B"0("'+D&"O*('"G&0PG"P&'A'"B&'";'GB"_!5
V$=
"
GD0A'"P+GR"\;`"7&'"0F'A+OO"_!5GD0A'"0C"B&'"')@'A*6'(BG"G&0P("*("G/;[,/A'"+R"\D`"7&'"_!5GD0A'"+B"
V$>
"
M*m'A'(B"]^_G"0C"B&'"')@'A*6'(BG"G&0P("*("G/;[,/A'"+R"\M`"]'(("M*+,A+6G"G&0P*(,"B&'"0F'AO+@"
V$K
"
0C"C+OG'"@0G*YF'"F+A*+(B"D+OOG";'BP''("1BA'O8+#-"h/B'DB#-"+(M"3O+*A1R"
V$L
"
"
V$Q
"
" "
V$V
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
Extended Data Figures
V=W
"
+"
V=!
"
"
V=#
"
"
V=$
"
;"
V==
"
"
V=>
"
"
V=K
"
Extended Data Figure 1. Performance differences between phasable and not-phasable
V=L
"
SNVs
V=Q
"
\+`"7&'"[,/A'"G&0PG"B&+B"+"G06+YD"F+A*+(B"/G/+OOH"0A*,*(+B'G"*("+"G*(,O'"G06+YD"D'OO"+(M"B&'("
V=V
"
G@A'+MG"B0"60A'"D'OOG"B&A0/,&"D'OO"M*F*G*0(-"A'G/OY(,"*("+"DO0(+O"D+AAH*(,"B&'"G+6'"F+A*+(BR"XB"+OG0"
V>W
"
G&0PG"&0P"B&'"6*G6+BD&'G"*("B&'"B/60A"G+6@O'"+(M"(0A6+O"G+6@O'"+A'"M*m'A'(B"CA06"'+D&"
V>!
"
0B&'AR"^"G06+YD"F+A*+(B"*G"60A'"O*8'OH"B0";'"+GG*,('M"B0"+"&+@O0BH@'"B&A0/,&"@&+G*(,-"P&*O'"+"
V>#
"
F+A*+(B"D+/G'M";H"A+(M06"G'N/'(D*(,"'AA0AG"*G"O'GG"O*8'OH"B0";'"G/DD'GGC/OOH"@&+G'MR"\;`"^"
V>$
"
@'AC0A6+(D'"D06@+A*G0("0C"G06+YD"F+A*+(BG"P&'A'"vqwT"D+(";'"@&+G'M-"+(M"vaqwT"D+((0BR"7&'"
V>=
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
[,/A'"G&0PG"+"&*,&'A"@'AC0A6+(D'"*("G06+YD"F+A*+(BG"B&+B"D+(";'"@&+G'M-"'G@'D*+OOH"+B"O0P'A"
V>>
"
]^_GR"9'"/G'M">We#>5C0OM"I33!$V>ed2"+(M"@A*0A*Yk'5A'D+OO"60M'R"
V>K
"
"
V>L
"
"
V>Q
"
+"
V>V
"
"
VKW
"
;"
VK!
"
"
VK#
"
"
VK$
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
Extended Data Figure 2. Visualization of neural network inputs.
VK=
"
\+`"q*O'/@5;+G'M"D+OO*(,"*(@/B"F*G/+O*k+Y0(R"7&'"D+(M*M+B'"G*B'"*G"D'(B'A'M"+(M"6+A8'M";H"BP0"
VK>
"
M+G&'M"O*('GR"\;`"_/OO5+O*,(6'(B5;+G'M"D+OO*(,"*(@/B"F*G/+O*k+Y0(R"X(";-"B&'"B0@"+(M";0g06"+A'"
VKK
"
@+MM'M"P*B&"k'A0"P&'("B&'"B0B+O"D0F'A+,'"0C"B/60A"+(M"(0A6+O"G+6@O'G"M0'G"(0B"A'+D&"B&'"
VKL
"
*(@/B"O*6*BR"7&'"(0A6+O"A'+M"+O*,(6'(BG"+(M"B/60A"A'+M"+O*,(6'(BG"*("+OO"D&+(('OG"+A'"
VKQ
"
G'@+A+B'M";H"BP0"A0PG"[OO'M"P*B&"k'A0GR"7&'"BP0"M'60(GBA+Y0(G"*(F0OF'M"BA/B&"F+A*+(BG"
VKV
"
A+(M06OH"@*D8'M"CA06"B&'"I33!$V>ed2"M+B+G'BR"
VLW
"
"
VL!
"
"
VL#
"
"
VL$
"
"
VL=
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint
Extended Data Figure 3. Two examples of haplotype inconsistency that signifies a false
VL>
"
somatic call.
VLK
"
\+`"S)+6@O'"0C"+"C+OG'"G06+YD"D+OO"P*B&"+"&+@O0BH@'"*(D0(G*GB'(B"P*B&"B&'"&+@O0BH@'"M'A*F'M"
VLL
"
CA06"+"&'B'A0kH,0/G",'A6O*('"F+A*+(B"('+A;HR"\;`"S)+6@O'"0C"+"C+OG'"G06+YD"D+OO"P*B&"+"
VLQ
"
&+@O0BH@'"*(D0(G*GB'(B"P*B&"B&'"&+@O0BH@'G"M'A*F'M"CA06"+"&060kH,0/G",'A6O*('"F+A*+(B"
VLV
"
('+A;HR"7&'";+G'G"^-"3-"l-"+(M"7"+A'"M'@*DB'M"*(",A''(-";O/'-"H'OO0P-"+(M"A'M-"A'G@'DYF'OHR"7&'"
VQW
"
;+D8,A0/(M"*(",A+H-"@/A@O'-"+(M"@*(8"A'@A'G'(BG"+("/(8(0P("&+@O0BH@'-"&+@O0BH@'"!-"+(M"
VQ!
"
&+@O0BH@'"#-"A'G@'DYF'OHR"l7"*G"+("+;;A'F*+Y0("0C",'(0BH@'R"
VQ#
"
"
VQ$
"
"
VQ=
"
.CC-BY-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2023. ; https://doi.org/10.1101/2023.08.17.553778doi: bioRxiv preprint