\A

B

\

#

Fig.#S9.#The#GC#correction#and#crossPsample#normalization#to#reduce#the#systematic#bias#generated#by#singlePcell#MALBAC.\A)\

19\of\33\

\

each\bin\of\single\cell\sequencing\data\through\MALBAC\is\biased.\The\copy\number\(right)\are\plotted\against\each\other.\MALBAC\shows\a\strong\correlation\between\correlation\in\MALBAC\is\sequence6dependent\while\the\bias\in\eMDA\is\less\sequence6dependent.\(B)\The\normalization\process\of\the\MALBAC\data.\The\sequence\depth\pattern\of\single\cell\MALBAC\is\correlated\with\GC\content\throughout\the\genome.\The\amplification\bias\is\corrected\by\LOWESS\GC\smoothing.\After\cross6sample\normalization,\the%unevenness\of\amplification\has\The\kb.\\\

\\\\\

\

\

20\of\33\

\

#

Fig.#S10.#Ploidy#measurement#of#HTP29#cells#using#fluorescencePactivated#cell#sorting#(FACS)#and#sequencing.\We%used\the\human\fibroblast\as\a\reference.\We%used\the\Hoechst\33342\(Invitrogen)\dye\to\stain\the\cells\and\applied\fluorescence\intensity.\(A)\DNA\content\distribution\of\fibroblasts.\(B)\DNA\content\distribution\of\HT629\cells.\(C)\DNA\content\distribution\of\the\~1:1\fibroblasts:\HT629\cells.\Comparing\to\the\fibroblasts,\normal\diploid\human\cells,\D)\distribution\peak\3.08\for\all\the%uniquely\mappable\regions.\The\difference\between\the\two\

21\of\33\

\

calculated\results\may\come\from\the\copy\number\difference\of\the\repeat\regions\or\the\two\cell\lines’\GC\content\difference,\which\affects\the\binding\ability\\\

\

\

22\of\33\