scieee AI-readable full text Open interactive document viewer

Discovery of new senolytics using machine learning

Smer-Barreto, Vanessa,Quintanilla, Andrea,Elliot, Richard J. R.,Dawson, John C.,Sun, Jiugeng,Carragher, Neil O.,Acosta, Juan C.,Oyarzún, Diego A.

Abstract

No

Full text

'LVFRYHU\RIQHZVHQRO\WLFVXVLQJPDFKLQHOHDUQLQJ 9DQHVVD 6PHU%DUUHWR $QGUHD 4XLQWDQLOOD 5LFKDUG - 5 (OOLRW -RKQ & 'DZVRQ -LXJHQJ 6XQ 1HLO2&DUUDJKHU-XDQ&DUORV$FRVWD'LHJR $2\DU]~Q &DQFHU5HVHDUFK8.(GLQEXUJK&HQWUH05&,QVWLWXWH RI*HQHWLFVDQG&DQFHU8QLYHUVLW\RI (GLQEXUJK&UHZH5RDG(GLQEXUJK(+;58. ,QVWLWXWRGH%LRPHGLFLQD\%LRWHFQRORJtDGH&DQWDEULD ,%%7(&&6,&8QLYHUVLGDGGH &DQWDEULD62'(5&$1&$OEHUW(LQVWHLQ6DQWDQGHU6SDLQ 6FKRRORI,QIRUPDWLFV8QLYHUVLW\RI(GLQEXUJK &ULFKWRQ6W(GLQEXUJK(+$%8. 6FKRRORI%LRORJLFDO6FLHQFHV8QLYHUVLW\RI(GLQEXUJK 0D[%RUQ&UHVFHQW(GLQEXUJK(+%) 8. 7KH$ODQ7XULQJ,QVWLWXWH(XVWRQ5RDG/RQGRQ 1:'%8. &RUUHVSRQGLQJDXWKRUVYDQHVVDVPHUEDUUHWR#HGDFXN MXDQDFRVWD#XQLFDQHV GR\DU]XQ#HGDFXN $EVWUDFW &HOOXODU VHQHVFHQFH LV D VWUHVV UHVSRQVH FKDUDFWHULVHG E\ D SHUPDQHQW FHOO F\FOH DUUHVW DQG D SURLQIODPPDWRU\ VHFUHWRPH ,Q DGGLWLRQ WR LWV WXPRXU VXSSUHVVRU UROH VHQHVFHQFH LV LQYROYHG LQ DJHLQJ DQG SURPRWHV PDQ\ GLVHDVH SURFHVVHV VXFK DV FDQFHU W\SH  GLDEHWHV RVWHRDUWKULWLV DQG 6$56&R9 LQIHFWLRQ 7KHUH LV D JURZLQJ LQWHUHVW LQ WKHUDSLHV EDVHG RQ WDUJHWHG HOLPLQDWLRQ RI VHQHVFHQW FHOOV \HW VR IDU RQO\ D IHZ VXFK VHQRO\WLFV DUH NQRZQ SDUWO\ GXH WR WKH SRRU JUDVS RI WKH PROHFXODU PHFKDQLVPV WKDW FRQWURO WKH VHQHVFHQFH VXUYLYDO SURJUDPPH +HUH ZH UHSRUW D KLJKO\ HIIHFWLYH PDFKLQH OHDUQLQJ SLSHOLQH IRU WKH GLVFRYHU\ RI VHQRO\WLF FRPSRXQGV 8VLQJ VROHO\ SXEOLVKHG GDWD ZH WUDLQHG PDFKLQH OHDUQLQJ DOJRULWKPV WR FODVVLI\ FRPSRXQGV DFFRUGLQJ WR WKHLU VHQRO\WLF DFWLRQ 0RGHOV ZHUH WUDLQHG RQ DV IHZ DV  NQRZQ VHQRO\WLFV DJDLQVW D EDFNJURXQG RI )'$DSSURYHG FRPSRXQGV RU LQ ODWHVWDJH FOLQLFDO GHYHORSPHQW  LQ WRWDO :H FRPSXWDWLRQDOO\ VFUHHQHG YDULRXV FKHPLFDO OLEUDULHV DQG VLQJOHG RXW WRS FDQGLGDWHV IRU YDOLGDWLRQ LQ KXPDQ OXQJ ILEUREODVWV ,05 DQG OXQJ DGHQRFDUFLQRPD $ FHOO OLQHV 7KLV OHG WR WKH GLVFRYHU\ RI WKUHH QRYHO VHQRO\WLFV JLQNJHWLQ ROHDQGULQ DQG SHULSORFLQ ZLWK SRWHQF\ FRPSDUDEOH WR FXUUHQW VHQRO\WLFV DQG D VHYHUDO KXQGUHGIROG UHGXFWLRQ LQ H[SHULPHQWDO VFUHHQLQJ FRVWV 2XU ZRUN GHPRQVWUDWHV WKDW PDFKLQH OHDUQLQJ FDQ WDNH PD[LPXP DGYDQWDJH RI H[LVWLQJ GUXJ VFUHHQLQJ GDWD SDYLQJ WKH ZD\ IRU QHZRSHQVFLHQFHDSSURDFKHVWRGUXJGLVFRYHU\IRUVHQHVFHQFHDVVRFLDWHGGLVHDVHV .H\ZRUGV VHQHVFHQFHPDFKLQHOHDUQLQJDUWLILFLDO LQWHOOLJHQFHGUXJGLVFRYHU\ .CC-BY-NC-ND 4.0 International licensemade available 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 The copyright holder for this preprintthis version posted April 27, 2022. ; https://doi.org/10.1101/2022.04.26.489505doi: bioRxiv preprint ,QWURGXFWLRQ 6HQHVFHQFH LV D FHOOXODU VWDWH FKDUDFWHULVHG E\ SHUPDQHQW FHOO F\FOH DUUHVW PDFURPROHFXODU GDPDJH DQG D UDQJH RI PHWDEROLF DOWHUDWLRQV >@ 7KH VHQHVFHQW SKHQRW\SH FDQ EH WULJJHUHG E\ GLIIHUHQW VWUHVVRUV VXFK DV UHSOLFDWLYH H[KDXVWLRQ FKHPRWKHUDSHXWLF DJHQWV RQFRJHQLF DFWLYDWLRQ DQG UDGLDWLRQ >@ 6HQHVFHQW FHOOV DIIHFW WLVVXH PLFURHQYLURQPHQW LQ PXOWLSOH ZD\V ERWK EHQHILFLDO DQG GHOHWHULRXV >@ )RU H[DPSOH VHQHVFHQFH DLGV PDPPDOLDQ HPEU\RQLF GHYHORSPHQW SURPRWHV ZRXQG KHDOLQJ DQG VWHPQHVV >@ DQG LV D SRWHQW WXPRXU VXSSUHVVLRQ PHFKDQLVP WKDW UHVWUDLQV WKH JURZWK RI FHOOV LQ GDQJHU RI PDOLJQDQW DOWHUDWLRQV >@ &RQYHUVHO\ VHQHVFHQW FHOOV DOVR SURPRWH WXPRULJHQHVLV DQG YDULRXV DJHUHODWHG PDOLJQDQFLHV GXH WR WKH VHFUHWLRQ RI D FRPSOH[ VHW RI SURWHLQV NQRZQ DV WKH VHQHVFHQFHDVVRFLDWHG VHFUHWRU\ SKHQRW\SH 6$63 >@ %HVLGHV WKHLU UROH LQ FDQFHU DQG DJHLQJ >@ WKH VHQHVFHQW SURJUDPPH KDV EHHQ OLQNHG WR DGYHUVH HIIHFWV LQ D EURDG UDQJH RI FRQGLWLRQV LQFOXGLQJ RVWHRSRURVLV RVWHRDUWKULWLV SXOPRQDU\ ILEURVLV 6$56&R9 LQIHFWLRQ KHSDWLF VWHDWRVLV DQG QHXURGHJHQHUDWLRQ >@ $V D UHVXOW WKHUH LV D JURZLQJ LQWHUHVW LQ WKH GLVFRYHU\ RI QHZ VHQRO\WLFV LHWKHUDSHXWLFDJHQWVWKDWVHOHFWLYHO\WDUJHWVHQHVFHQWFHOOVIRUHOLPLQDWLRQ >@ 6HQRO\WLFV KDYH VKRZQ VXEVWDQWLDO SURPLVH LQ DPHOLRUDWLQJ V\PSWRPV RI PDQ\ FRQGLWLRQV LQ PLFH >±@ 'HVSLWH WKHVH HQFRXUDJLQJ UHVXOWV WR GDWH WKHUH DUH IHZ NQRZQ FRPSRXQGV ZLWK SURYHQ VHQRO\WLF DFWLRQ >±±@ DQG RQO\ WZR FRPSRXQGV KDYH VKRZQ HIILFDF\ LQ FOLQLFDO WULDOV GDVDWLQLE DQG TXHUFHWLQ LQ FRPELQDWLRQ WKHUDS\ >@ 6RPH RI WKH PRVW VFUXWLQLVHG VHQRO\WLFV ZHUH LGHQWLILHG E\ WDUJHWLQJ DQWLDSRSWRWLF SURWHLQV XSUHJXODWHG LQ VHQHVFHQFH VXFK DV WKH %FO IDPLO\ LQKLELWRUV QDYLWRFOD[ >@ DQG $%7 >@ 2WKHU VHQRO\WLFV ZHUH GLVFRYHUHG WKURXJK SDQHO VFUHHQV >@ DQG PRUH UHFHQWO\ KLJKWKURXJKSXW VFUHHQV KDYH LGHQWLILHG FDUGLDF JO\FRVLGHV VXFK DV RXDEDLQ >@ DQG GLJR[LQ >@ DV SRWHQW VHQRO\WLF DJHQWV $ NH\ FKDOOHQJH IRU VHQRO\WLF WKHUDSLHV WR VXFFHHG LV WKDW PDQ\ VXFK FRPSRXQGV GLVSOD\ FHOOW\SH VSHFLILF DFWLRQ ,Q DGGLWLRQ FHUWDLQ VHQRO\WLFV WKDW ZRUN ZHOO IRU RQH FHOOW\SH DUH KLJKO\ WR[LF DJDLQVW RWKHU QRQVHQHVFHQW FHOOW\SHV >@ ,Q WKH FDVH RI FDQFHU WKHUDSLHV PRVW VHQRO\WLFV WDUJHW SDWKZD\V WKDW DUH PXWDWHG LQ FDQFHU OLPLWLQJ WKHLU DSSOLFDELOLW\ DV WKHUDSHXWLF DJHQWV >@ $V D UHVXOW WKHUH LV D QHHG IRU QHZ VWUDWHJLHV WR LGHQWLI\ FRPSRXQGVZLWKVHQRO\WLFDFWLRQ &RPSXWDWLRQDO VFUHHQV KDYH EHHQ HPSOR\HG IRU GHFDGHV LQ GUXJ GLVFRYHU\ >±@ $UWLILFLDO ,QWHOOLJHQFH $, PHWKRGV LQ SDUWLFXODU KDYH EHHQ ZLGHO\ DGRSWHG E\ LQGXVWULDO DQG DFDGHPLF ODERUDWRULHV GXH WR WKHLU DELOLW\ WR GHWHFW KLGGHQ SDWWHUQV LQ ODUJH FROOHFWLRQV RI FKHPLFDO GDWD ZKLFK FDQ EH XVHG WR QDUURZ GRZQ WKH VHDUFK VSDFH IRU DFWLYH PROHFXOHV 7KHVH DSSURDFKHV W\SLFDOO\ UHO\ RQ D FRPELQDWLRQ RI PROHFXODU G\QDPLFV DQG VRSKLVWLFDWHG SLSHOLQHV WR QDYLJDWH WKH VSDFH RI GUXJ FDQGLGDWHV >@ 7KDQNV LQ SDUW WR WKH GHOXJH RI PROHFXODU GDWD DYDLODEOH UHFHQW \HDUV KDYH ZLWQHVVHG WKH DGRSWLRQ RI PDFKLQH OHDUQLQJ PHWKRGV WUDLQHG RQ PROHFXODU ILQJHUSULQWV RU OHDUQHG UHSUHVHQWDWLRQV RI FKHPLFDO VWUXFWXUHV WDVNHG ZLWK SUHGLFWLQJ PROHFXODU DFWLYLW\ LQ D YDULHW\ RI ELRORJLFDO FRQWH[WV >±@ 6HYHUDO RI WKHVH PHWKRGV GHSDUW IURP WUDGLWLRQDO WDUJHWRULHQWHG DSSURDFKHV WR GUXJ GLVFRYHU\ LQ IDYRXU RI WDUJHWDJQRVWLF VWUDWHJLHV >@ WKDW HPSOR\ YDULRXV SKHQRW\SLF PHDVXUHPHQWV IRU PRGHO WUDLQLQJ 6XFK WDUJHWDJQRVWLF DSSURDFK RIIHUV QHZ DYHQXHV WR H[SDQG WKH UDQJH RI FKHPLFDO VWDUWLQJ SRLQWV LQ HDUO\ SKDVHV RI WKH GUXJ GLVFRYHU\ SLSHOLQH >@ DQG LV SDUWLFXODUO\ ZHOOVXLWHG IRU VHQRO\WLFV GLVFRYHU\ JLYHQ WKH UHODWLYHO\ SRRU JUDVS RI WKH PROHFXODU SDWKZD\VWKDWFRQWUROWKHVHQHVFHQWSKHQRW\SH +HUH ZH UHSRUW WKH GHYHORSPHQW DQG YDOLGDWLRQ RI D PDFKLQH OHDUQLQJ SLSHOLQH IRU WKH GLVFRYHU\ RI VHQRO\WLFV :H DVVHPEOHG D GDWDVHW PLQHG IURP PXOWLSOH VRXUFHV >±±@ LQFOXGLQJ DFDGHPLF SXEOLFDWLRQV DQG D FRPPHUFLDO SDWHQW DQG HPSOR\HG LW WR WUDLQ PDFKLQH PRGHOV SUHGLFWLYH RI VHQRO\WLF DFWLRQ :H FRPSXWDWLRQDO VFUHHQHG D OLEUDU\ RI PRUH WKDQ  FRPSRXQGV DQG LGHQWLILHG D UHGXFHG .CC-BY-NC-ND 4.0 International licensemade available 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 The copyright holder for this preprintthis version posted April 27, 2022. ; https://doi.org/10.1101/2022.04.26.489505doi: bioRxiv preprint VHW RI  KLWV IRU H[SHULPHQWDO YDOLGDWLRQ 2XU H[SHULPHQWDO VFUHHQ LQ WZR PRGHO FHOO OLQHV RI RQFRJHQH DQG WKHUDS\LQGXFHG VHQHVFHQFH UHYHDOHG VHQRO\WLF DFWLYLW\ RI WKUHH FRPSRXQGV JLQNJHWLQ ROHDQGULQ DQG SHULSORFLQ ZLWK SRWHQFLHV DQG GRVHUHVSRQVHV FRPSDUDEOH WR NQRZQ VHQRO\WLFV 7KLV DSSURDFK GHPRQVWUDWHV WKDW FRPPXQLW\ DQG RSHQ GDWD FDQ EH PD[LPDOO\ OHYHUDJHG WR GHWHFW QHZ DFWLYH WKHUDSHXWLF FRPSRXQGV DQG WKXV VHWV WKH PHWKRGRORJLFDO JURXQGZRUN IRU D QHZ RSHQVFLHQFHDSSURDFKWRGUXJGLVFRYHU\DQGUHSXUSRVLQJ 5HVXOWV 'DWDDVVHPEO\DQGTXDOLW\FRQWURO :H ILUVW DVVHPEOHG D GDWDVHW RI VHQRO\WLFV SRVLWLYHV DQG QRQVHQRO\WLFV QHJDWLYHV WR EH XVHG IRU PRGHO WUDLQLQJ )LJ $ 7R WKLV HQG ZH PLQHG D SDQHO RI  VHQRO\WLFV UHSRUWHG LQ WKH OLWHUDWXUH LQFOXGLQJ FRPSRXQGV IURP YDULRXV FKHPLFDO IDPLOLHV VXFK DV IODYRQRLGV FDUGLDF JO\FRVLGHV DQG DQWLELRWLFV VKRZQ WR GLVSOD\ VHQRO\WLF DFWLRQ 7KHVH SRVLWLYHV KDYH EHHQ VKRZQ WR SUHIHUHQWLDOO\ WDUJHW WKH VHQHVFHQW SKHQRW\SH LQ D YDULHW\ RI FHOO W\SHV )LJ % 6RPH RI WKH FRPSRXQGV HJ RXDEDLQ >@ KDYH EHHQ VKRZQ WR KDYH ZLGHVSHFWUXP VHQRO\WLF DFWLRQ ZKHUHDV RWKHUV VXFK DV %,; OLPLW WKHLU HIIHFW WR VSHFLILF FRQGLWLRQV >@ 7KH OLWHUDWXUH VRXUFHV XVHG WR DVVHPEOH WKH OLVW RI  VHQRO\WLFV XVHG GLIIHUHQW FHOO W\SHV VFUHHQLQJ DVVD\V DQG PHWKRGV IRU LQGXFWLRQ RI WKH VHQHVFHQW SKHQRW\SH :H FRPELQHG WKLV OLVW ZLWK DQRWKHU SDQHO RI  VHQRO\WLFV UHSRUWHG LQ D FRPPHUFLDO SDWHQW >@ 7KH SRVLWLYHV ZHUH WKHQ PHUJHG ZLWK D ODUJH EDFNJURXQG RI FRPSRXQGV DVVXPHG WR ODFN VHQRO\WLF DFWLRQ 7KLV DVVXPSWLRQ ZDV QHHGHG GXH WR WKH ODFN RI GDWD RQ QHJDWLYH VFUHHQLQJ UHVXOWV IURP SXEOLVKHG VRXUFHV 6LQFH PDFKLQH OHDUQLQJ PRGHOV FDQ ELDV WKHLU SUHGLFWLRQV WRZDUGV WKH WUDLQLQJ GDWD WKH\ KDYH EHHQ H[SRVHG WR ZH FKRVH WKH QHJDWLYHV IURP WZR GLYHUVH FKHPLFDO OLEUDULHV /23$& DQG 3UHVWZLFN )'$DSSURYHG ZKLFK FRQWDLQ D ZLGH UDQJH RI )'$DSSURYHG RU FOLQLFDOVWDJHFRPSRXQGV 7KH IXOO GDWDVHW IRU PRGHO WUDLQLQJ FRQWDLQV  FRPSRXQGV LQFOXGLQJ  SRVLWLYHV  :H GHOLEHUDWHO\ FKRVH WR RYHUUHSUHVHQW WKH QHJDWLYHV LQ WKH WUDLQLQJ GDWD VR DV WR UHIOHFW WKH ORZ OLNHOLKRRG RI D FKHPLFDO VWUXFWXUH EHLQJ D VHQRO\WLF )LJ $ 7R FRQYHUW WKH FKHPLFDO VWUXFWXUHV LQWR D QXPHULFDO IRUPDW WKDW FDQ EH XVHG IRU PRGHO WUDLQLQJ ZH ELQDUL]HG HDFK FRPSRXQG LQ WKH WUDLQLQJ OLEUDU\ DV  QHJDWLYH RU  SRVLWLYH DQG FRPSXWHG  SK\VLFRFKHPLFDO GHVFULSWRUV ZLWK WKH 5'.LW SDFNDJH >@ 7KHVH GHVFULSWRUV LQFOXGH EDVLF PROHFXODU SURSHUWLHV VXFK DV PD[LPXP SDUWLDO FKDUJH PROHFXODU ZHLJKW DQG QXPEHU RI YDOHQFH HOHFWURQV DV ZHOO DV VWUXFWXUDO SURSHUWLHV VXFK DV WKH PROHFXODUFRQQHFWLYLW\&KLLQGH[HV(6WDWHWRSRORJLFDOSDUDPHWHUVDQG.DSSDVKDSHLQGH[HV 7KH SRVLWLYH FRPSRXQGV ZHUH PLQHG IURP KLJKO\ KHWHURJHQHRXV VRXUFHV WKDW XWLOLVHG GLIIHUHQW FHOO OLQHV DQG VFUHHQLQJ DVVD\V )LJ % 7KLV EHDUV WKH ULVN RI LQWURGXFLQJ ELDV LQ RXU PRGHOV DQG OLPLW WKHLU SUHGLFWLYH SRZHU LI VSHFLILF FKHPLFDO VWUXFWXUHV RU FODVVHV DUH RYHUUHSUHVHQWHG LQ WKH WUDLQLQJ GDWD 7KLV LV IXUWKHU H[DFHUEDWHG E\ WKH KHDY\ LPEDODQFH EHWZHHQ WKH QXPEHU RI SRVLWLYHV DQG QHJDWLYHV LQFOXGHG LQ WKH WUDLQLQJ GDWD :H WKXV VRXJKW WR FDUHIXOO\ TXDQWLI\ WKH GLYHUVLW\ RI WKH  SRVLWLYHV XVLQJ WKH 5'.LW GHVFULSWRUV DV IHDWXUH YHFWRUV DVVRFLDWHG ZLWK HDFK FRPSRXQG 7R DVVHVV GLYHUVLW\ RI WKH WUDLQLQJ GDWD ZH H[DPLQHG WKH FOXVWHU VWUXFWXUH RI WKH SRVLWLYH FRPSRXQGV XVLQJ WKUHH GLIIHUHQW PHWKRGRORJLHV )LJ &( :H ILUVW FOXVWHUHG WKH SRVLWLYHV ZLWK WKH NPHDQV DOJRULWKP >@ DQG WKH FRVLQH GLVWDQFH EHWZHHQ IHDWXUH YHFWRUV WKLV DQDO\VLV UHYHDOHG DQ DOPRVW OLQHDU GHFUHDVH LQ WKH NPHDQV VFRUH ZLWK UHVSHFW WR WKH QXPEHU RI FOXVWHUV )LJ & 7KH ODFN RI D FOHDU ³HOERZ´ LQ WKH NPHDQV VFRUH VXJJHVWV SRRU GDWD FOXVWHULQJ DQG KHQFH SURYLGHV D TXDOLWDWLYH LQGLFDWLRQ RI GLYHUVLW\ LQ WKH WUDLQLQJ GDWD 7R GHWHUPLQH WKH TXDOLW\ DQG FRQVLVWHQF\ RI WKHVH FOXVWHUV ZH FRPSXWHG WKH VLOKRXHWWH FRHIILFLHQWV IRU DOO FRPSRXQGV DQG QXPEHU RI FOXVWHUV N >@ ZH IRXQG FRQVLVWHQWO\ ORZ .CC-BY-NC-ND 4.0 International licensemade available 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 The copyright holder for this preprintthis version posted April 27, 2022. ; https://doi.org/10.1101/2022.04.26.489505doi: bioRxiv preprint YDOXHV IRU WKH VLOKRXHWWH FRHIILFLHQW DYHUDJHG DFURVV DOO FRPSRXQGV ZKLFK IXUWKHU VXJJHVWV OLWWOH VLPLODULW\DPRQJWKHVHQRO\WLFFRPSRXQGVFKRVHQIRUWUDLQLQJ $V D VHSDUDWH FKHFN IRU WKH GLYHUVLW\ RI WKH WUDLQLQJ GDWD ZH EXLOW WKH 7DQLPRWR GLVWDQFH JUDSK IRU DOO VHQRO\WLFV HPSOR\HG IRU WUDLQLQJ DQG ODEHOOHG HDFK FRPSRXQG DFFRUGLQJ WR WKH VRXUFH IURP ZKLFK WKH\ ZHUH REWDLQHG )LJ ' 1RGHV LQ WKH 7DQLPRWR GLVWDQFH JUDSK UHSUHVHQW FRPSRXQGV DQG WZR FRPSRXQGV DUH FRQQHFWHG E\ DQ HGJH LI WKH\ DUH VXIILFLHQWO\ FORVH LQ WKH 5'NLW FKHPLFDO GHVFULSWRU VSDFH 7KH VWUXFWXUH RI WKH UHVXOWLQJ GLVWDQFH JUDSK FRUURERUDWHV WKH ILQGLQJ WKDW PRVW VHQRO\WLFV DUH IDU DSDUW LQ WKH GHVFULSWRU VSDFH PHGLDQ 7DQLPRWR GLVWDQFH  )LJ ' LQVHW DQG WKXV WHQG WR EH KLJKO\ GLVVLPLODU WR HDFK RWKHU $V D ILQDO FKHFN ZH FOXVWHUHG WKH 7DQLPRWR GLVWDQFH JUDSK XVLQJ FRPPXQLW\ GHWHFWLRQ D W\SH RI FOXVWHULQJ WHFKQLTXH IURP QHWZRUN VFLHQFH >@ WKDW GRHV QRW UHTXLUH D SULRUL VSHFLILFDWLRQ RI WKH QXPEHU RI FOXVWHUV :H HPSOR\HG WKH SRSXODU /RXYDLQ DOJRULWKP >@ EHFDXVH RI LWV FRPSXWDWLRQDO HIILFLHQF\ DQG WKH LQFOXVLRQ RI D UHVROXWLRQ SDUDPHWHU Ȗ IRU WXQLQJ WKH JUDQXODULW\ RI WKH UHVXOWLQJ FOXVWHUV ODUJHU YDOXHV RI Ȗ OHDG WR PRUH FOXVWHUV DQG KHQFH D PRUH JUDQXODU SDUWLWLRQ RI WKH JUDSK )RU D ZLGH VZHHS RI WKH UHVROXWLRQ SDUDPHWHU ZH IRXQG DQ DOPRVW OLQHDU LQFUHDVH LQ WKH QXPEHU RI FOXVWHUV DQG WZR SODWHDXV DW N  DQG N  FOXVWHUV )LJ ( VXFK SODWHDXV VXJJHVW WKDW WKH GDWD PD\ QDWXUDOO\ FOXVWHU LQWR ILYH RU VL[ JURXSV :H IXUWKHU LQYHVWLJDWHG WKHVH FOXVWHUV DQG UHDVRQHG WKDW FRPSRXQGV PD\ DJJUHJDWH DFFRUGLQJ WR WKH VRXUFH ZKHUH WKH\ ZHUH PLQHG IURP 7R WHVW WKLV K\SRWKHVLV ZH TXDQWLILHG WKH VLPLODULW\ EHWZHHQ WKH /RXYDLQ FOXVWHUV DQG OLWHUDWXUH ODEHOV )LJ % XVLQJ WKH DGMXVWHG 5DQG LQGH[ $5, D XVHIXO PHWULF IRU FRPSDULQJ GLIIHUHQW FOXVWHUV WKDW FRUUHFWV IRU UDQGRP JURXS DVVLJQPHQWV >@ :H IRXQG H[WUHPHO\ ORZ $5, VFRUHV )LJ ( DFURVV DOO FOXVWHU UHVROXWLRQV GHVFULEHG E\ WKH Ȗ SDUDPHWHU PRUHRYHU WKH $5, VFRUHV VKRZHG SURQRXQFHG WURXJKV DW WKH SODWHDXV GHWHFWHG ZLWK WKH /RXYDLQ PHWKRG PHDQ $5, IRU  UXQV RI WKH FOXVWHULQJ PHWKRG ZKLFK ZH UHJDUGHG DV VXIILFLHQW HYLGHQFH WKDW FRPSRXQGV GR QRW FOXVWHUDFFRUGLQJWRWKHVRXUFHIURPZKLFKWKH\ZHUHREWDLQHG 3UHGLFWLQJVHQRO\WLFFRPSRXQGVE\FRPSXWDWLRQDOVFUHHQXVLQJPDFKLQHOHDUQLQJ :H QH[W VRXJKW WR WUDLQ PDFKLQH OHDUQLQJ PRGHOV RQ WKH DVVHPEOHG GDWDVHW ZLWK WKH DLP RI XVLQJ WKHP WR FRPSXWDWLRQDOO\ VFUHHQ FKHPLFDO OLEUDULHV DQG LGHQWLI\ KLWV IRU H[SHULPHQWDO YDOLGDWLRQ )LJ $ $ NH\ FKDOOHQJH IRU PRGHO WUDLQLQJ LV WKH KHDY\ LPEDODQFH EHWZHHQ WKH QXPEHU RI VHQRO\WLF DQG QRQVHQRO\WLF FRPSRXQGV :H WKXV VFRUHG WKH PRGHOV ZLWK D VXLWH RI SHUIRUPDQFH PHWULFV DFFXUDF\ IUDFWLRQ RI DOO FRPSRXQGV FRUUHFWO\ LGHQWLILHG SUHFLVLRQ IUDFWLRQ RI WUXH SRVLWLYH LGHQWLILFDWLRQV RXW RI DOO SRVLWLYH LGHQWLILFDWLRQV DQG UHFDOO IUDFWLRQ RI FRUUHFW LGHQWLILFDWLRQV RI WUXH SRVLWLYHV 7KH SUHFLVLRQ DQG UHFDOO PHWULFV DUH RI SDUDPRXQW LPSRUWDQFH LQ RXU SUREOHP EHFDXVH WKH FODVV LPEDODQFH WHQGV WR SURGXFH RYHURSWLPLVWLF DFFXUDF\ DV D UHVXOW RI FRUUHFW FODVVLILFDWLRQ RI WKH PDMRULW\ FODVVHYHQZKHQWKHPLQRULW\FODVVLVSRRUO\FODVVLILHG >@ :H IRFXVHG SULPDULO\ RQ WZR FRPPRQ PRGHOV IRU ELQDU\ FODVVLILFDWLRQ VXSSRUW YHFWRU PDFKLQHV 690 DQG UDQGRP IRUHVWV 5) 7KHVH PRGHOV RSHUDWH E\ SDUWLWLRQLQJ WKH IHDWXUH VSDFH VR DV WR HQVXUH WKDW SRVLWLYH DQG QHJDWLYH VDPSOHV DUH RSWLPDOO\ DVVLJQHG WR D SDUWLWLRQ ,Q WKHLU EDVLF IRUP 690 VOLFHV WKH IHDWXUH VSDFH ZLWK D K\SHUSODQH ZKLOH 5) DUH HQVHPEOHV RI GHFLVLRQ WUHHV WKDW VHJPHQW WKH IHDWXUH VSDFH ZLWK RUWKRJRQDO FXWV DFURVV HDFK IHDWXUH >@ 0RGHOV ZHUH WUDLQHG XVLQJ JUDGLHQWEDVHG RSWLPL]DWLRQ RQ D UHGXFHG VHW RI  QRUPDOLVHG 5'.LW IHDWXUHV WKDW ZHUH LGHQWLILHG ZLWK IHDWXUH LPSRUWDQFH DQDO\VLV 6XSSOHPHQWDU\ )LJ $ DQG XVLQJ NIROG FURVVYDOLGDWLRQ WR DYRLG RYHUILWWLQJDQGUREXVWPRGHOFRPSDULVRQ :H IRXQG WKDW ERWK 690 DQG 5) PRGHOV GLVSOD\HG SRRU SHUIRUPDQFH )LJ % EXW VKRZHG PDUNHG GLIIHUHQFHV LQ WKH W\SH RI PLVFODVVLILFDWLRQ HUURUV WKH\ SURGXFH 7KH FRQIXVLRQ PDWULFHV )LJ % .CC-BY-NC-ND 4.0 International licensemade available 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 The copyright holder for this preprintthis version posted April 27, 2022. ; https://doi.org/10.1101/2022.04.26.489505doi: bioRxiv preprint VXJJHVW WKDW WKH 5) PRGHO WHQGV WR UHWXUQ IHZ IDOVH SRVLWLYHV KLJK SUHFLVLRQ DQG D KLJK QXPEHU RI IDOVH QHJDWLYHV ORZ UHFDOO ZKHUHDV WKH 690 UHWXUQV RSSRVLWH UHVXOWV :H REVHUYHG VLPLODU WUDGHRIIV EHWZHHQ IDOVH QHJDWLYHV DQG IDOVH SRVLWLYHV LQ YDULRXV RWKHU PRGHOV RI YDULHG FRPSOH[LW\ 6XSSOHPHQWDU\ 7DEOH  LQFOXGLQJ VLPSOH ORJLVWLF UHJUHVVRUV %D\HVLDQ FODVVLILHUV DQG GHHS OHDUQLQJ PRGHOV >@ ,Q OLQH ZLWK H[SHFWDWLRQ WKH PDMRULW\ RI PRGHOV GLVSOD\HG KLJK DFFXUDF\ ! EXW WKLV LV DQ DUWHIDFW UHVXOWLQJ IURP WKH FRUUHFW FODVVLILFDWLRQ RI PDQ\ QRQVHQRO\WLF FRPSRXQGV WKDW DUH RYHUUHSUHVHQWHGLQWKHWUDLQLQJGDWD )RU WKH SXUSRVHV RI HDUO\VWDJH GUXJ GLVFRYHU\ IDOVH SRVLWLYHV DUH PRUH GHOHWHULRXV WKDQ IDOVH QHJDWLYHV EHFDXVH WKH\ DUWLILFLDOO\ LQIODWH WKH QXPEHU RI SUHGLFWHG KLWV DQG WKXV LQFUHDVH WKH FRVWV RI GRZQVWUHDP H[SHULPHQWDO YDOLGDWLRQ :H WKXV WRRN WKH SHUIRUPDQFH RI WKH 5) PRGHO DV D EDVHOLQH DQG DLPHG DW LPSURYLQJ LWV SUHGLFWLYH SRZHU E\ XVLQJ DQ HQVHPEOH PRGHO ;*%RRVW WKDW LV NQRZQ WR LPSURYH SHUIRUPDQFH E\ DJJUHJDWLQJ SUHGLFWLRQV IURP D FROOHFWLRQ RI GHFLVLRQ WUHHV >@ 7KH ;*%RRVW PRGHO LPSURYHG ERWK SUHFLVLRQ DQG UHFDOO VFRUHV DQG RYHUDOO UHWXUQHG WKH EHVW SHUIRUPDQFH DPRQJ DOO FRQVLGHUHG PRGHOV )LJ % :H REVHUYHG SUHFLVLRQ VFRUHV RI XS WR  LQ VRPH RI WKH FURVVYDOLGDWLRQ UXQV ZKLFK DPRXQWV WR D IDOVH GLVFRYHU\ UDWH RI  WKDW ZH UHJDUGHG DV DFFHSWDEOH JLYHQWKHKHWHURJHQHLW\RIWKHGDWDHPSOR\HGIRUWUDLQLQJ :H QH[W HPSOR\HG WKH SUHWUDLQHG ;*%RRVW PRGHO WR FRPSXWDWLRQDOO\ VFUHHQ D OLEUDU\ RI FKHPLFDO VWUXFWXUHV VHOHFWHG RQ WKH EDVLV RI FKHPLFDO GLYHUVLW\ :H VFUHHQHG D OLVW RI FRPSRXQGV FRQVLVWLQJ RI WKH FRQWHQWV IURP WKH / 7DUJHW0RO $QWLFDQFHU DQG / 6HOOHFN )'$DSSURYHG  3DVVHG 3KDVH OLEUDULHV LQWR D VLQJOH GDWDVHW ZLWK  FRPSRXQGV IHDWXULVHG ZLWK WKH SK\VLFRFKHPLFDO GHVFULSWRUV IURP 5'.LW QRQH RI WKH FRPSRXQGV LQ WKH VFUHHQLQJ OLEUDU\ ZHUH HPSOR\HG IRU WUDLQLQJ 7KH FRPSXWDWLRQDO VFUHHQ UHYHDOHG WKH ;*%RRVW PRGHO WR EH H[FHSWLRQDOO\ VHOHFWLYH )LJ & ZLWK WKH YDVW PDMRULW\ RI FRPSRXQGV KDYLQJ H[WUHPHO\ ORZ SUHGLFWLRQ VFRUHV DQG WKXV GHHPHG WR KDYH D ORZ SUREDELOLW\ RI EHLQJ VHQRO\WLF ,PSRUWDQWO\ WKH VFRUH GLVWULEXWLRQ )LJ & GLVSOD\HG D VPDOO JURXS RI  FRPSRXQGV  RI WKH IXOO OLEUDU\ ZLWK D FRPSDUDWLYHO\ KLJKHU SUREDELOLW\ RI EHLQJ VHQRO\WLF )LJ & RUDQJH WKLV VPDOO VHW RI FRPSRXQGV ZHUH VFRUHG DERYH  E\ RXU ;*%RRVW PRGHO DQG ZHUH VHOHFWHG IRU IXUWKHU H[SHULPHQWDO YDOLGDWLRQ :H QRWH WKDW WKH VHOHFWHG FRPSRXQGV DUH H[WUHPH RXWOLHUV ZLWK SUHGLFWLRQ VFRUHV DW OHDVW  VWDQGDUG GHYLDWLRQV DZD\ IURP WKH EXON RI FRPSRXQGV VFUHHQHG )LJ ' :H HPSOR\HG GLPHQVLRQDOLW\ UHGXFWLRQ WR YLVXDOLVH WKH WUDLQLQJ DQG VFUHHQLQJ FRPSRXQGV LQ WKH 5'.LW IHDWXUH VSDFH >@ ZKLFK UHYHDOHG D VWURQJ RYHUODS EHWZHHQ WKH WZR VHWV DQG WKXV D VWURQJ HYLGHQFH WKDW WKH FRPSXWDWLRQDO VFUHHQ ZDV SHUIRUPHG LQ WKH KLJKFRQILGHQFH GRPDLQRIWKHPDFKLQHOHDUQLQJPRGHO)LJ( ,GHQWLILFDWLRQRIVHQRO\WLFVE\H[SHULPHQWDOVFUHHQLQJRIWRSSUHGLFWHGFRPSRXQGV 7R YDOLGDWH RXU PDFKLQH OHDUQLQJ DSSURDFK ZH H[SHULPHQWDOO\ VFUHHQHG WKH WRS SUHGLFWHG PROHFXOHV )LJ ' IRU VHQRO\WLF DFWLYLW\ LQ WZR PRGHO FHOO OLQHV IRU RQFRJHQHLQGXFHG DQG WKHUDS\LQGXFHG VHQHVFHQFH :H ILUVW DVVHVVHG RQFRJHQHLQGXFHG VHQHVFHQFH 2,6 LQ KXPDQ GLSORLG ILEUREODVWV ,05 WUDQVGXFHG ZLWK WKH IXVLRQ SURWHLQ (55$6 ,05 (55$6 ZKLFK LQGXFHV RQFRJHQLF 5DV*9PHGLDWHG VWUHVV E\ DGGLWLRQ RI K\GUR[\WDPR[LIHQ 2+7 WR WKH FXOWXUH PHGLD >@ 7UHDWPHQW RI ,05 (55$6 FHOOV ZLWK 2+7 VKRZHG D GHFUHDVH LQ SUROLIHUDWLRQ LQFUHDVHG VHQHVFHQFH DVVRFLDWHG ȕJDODFWRVLGDVH DFWLYLW\ LQGXFWLRQ RI FHOO F\FOH LQKLELWRU H[SUHVVLRQ DQG DFWLYDWLRQ RI WKH 6$63 ZKHQ FRPSDUHG ZLWK FRQWURO DQG 2+7 XQWUHDWHG FHOOV LQGLFDWLQJ WKDW WKH FHOOVXQGHUZHQWRQFRJHQHLQGXFHGVHQHVFHQFH6XSSOHPHQWDU\)LJ 7R WHVW WKH VHQRO\WLF DFWLYLW\ ZH FRPSDUHG WKH HIIHFW RI HDFK FRPSRXQG RQ WKH WRWDO FHOO QXPEHU DXWRPDWHG KLJK FRQWHQW LPDJHEDVHG DQDO\VLV RI WRWDO QXPEHU RI QXFOHL SHU ZHOO LQ QRQVHQHVFHQW .CC-BY-NC-ND 4.0 International licensemade available 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 The copyright holder for this preprintthis version posted April 27, 2022. ; https://doi.org/10.1101/2022.04.26.489505doi: bioRxiv preprint DQG VHQHVFHQW ,05 (55$6 FXOWXUHV WUHDWHG ZLWK WKH WRS KLWV IURP WKH FRPSXWDWLRQDO VFUHHQ )LJ ' )LJ $' 6XSSOHPHQWDU\ )LJ $& 7KH GURS LQ FHOO QXPEHU FRPSDUHG WR XQWUHDWHG FRQWUROV LV LQGLFDWHG E\ WKH QXFOHL FRXQW DQG UHIOHFWV FHOO GHDWK >@ $V D SRVLWLYH FRQWURO ZH XVHG RXDEDLQ D FDUGLDF JO\FRVLGH ZLWK ZHOO FKDUDFWHULVHG VHQRO\WLF DFWLYLW\ >@ $Q RSWLPDO VHQRO\WLF HIIHFW ZDV UHDFKHG ZLWK DGGLWLRQ RI  Q0 RXDEDLQ ,& FRQWURO  Q0 ,& VHQHVFHQFH  Q0 WR 2+7LQGXFHG ,05 (55$6 FHOOV 7KLV FRQFHQWUDWLRQ NLOOHG PRVW RI WKH FXOWXUHG VHQHVFHQW FHOOV EXWUHVXOWHGLQQRUHGXFWLRQLQQXPEHUVRIQRQVHQHVFHQWFRQWUROFHOOV)LJ% :H WKHQ WHVWHG WKH  FDQGLGDWH FRPSRXQGV DQG IRXQG WKUHH ZLWK FOHDU VHQRO\WLF DFWLRQ SHULSORFLQ DQG ROHDQGULQ WZR FDUGLDF JO\FRVLGHV ZKLFK KDYH QRW EHHQ SUHYLRXVO\ LGHQWLILHG DV VHQRO\WLFV DQG JLQNJHWLQ D QDWXUDO QRQWR[LF ELIODYRQH WKLV DPRXQWV WR D KLW FRQILUPDWLRQ UDWH RI  )LJ &' 6XSSOHPHQWDU\ )LJ & 7UHDWPHQW RI VHQHVFHQW ,05 (55$6 FHOOV ZLWK WKH WKUHH FRPSRXQGV VKRZHG UHGXFHG QXFOHL FRXQWV ZKHQ FRPSDUHG ZLWK SUROLIHUDWLQJ QRQVHQHVFHQW ,05 FRQWUROV ZLWK DQ HIIHFW FRPSDUDEOH WR WKH SRVLWLYH FRQWURO )LJ &' 6XSSOHPHQWDU\ )LJ $ )RU FRQILUPDWLRQ ZH SHUIRUPHG FHOO VWDLQLQJ ZLWK +RHFKVW WR ODEHO WKH QXFOHL VKRZLQJ WKH FOHDUHVW HIIHFW ZLWK GRVHV RI  Q0 ROHDQGULQ ,& FRQWURO  Q0 ,& VHQHVFHQFH  Q0  Q0 SHULSORFLQ ,& FRQWURO  Q0 ,& VHQHVFHQFH  Q0 DQG  ȝ0 JLQNJHWLQ ,& FRQWURO  ȝȂ ,& VHQHVFHQFH  ȝȂ 7KHVH FRQFHQWUDWLRQV KDYH D PDUJLQDO HIIHFW RQ QRUPDO FHOOV¶ QXFOHL FRXQWV EXW D PDUNHG GHFUHDVH LQ QXFOHL FRXQWV LQ VHQHVFHQW FHOOV )LJ &' 6XSSOHPHQWDU\ )LJ $ 7KH GRVHUHVSRQVH FXUYHV VKRZHG D ORZHU ,& IRU WKH WKUHH FRPSRXQGV LQ QXFOHL FRXQWV IURP VHQHVFHQW FHOOV DV FRPSDUHG WR QXFOHL IURP QRUPDO FHOOV )LJ ' ,Q SDUWLFXODU SHULSORFLQ WKH WRS KLW SUHGLFWHG E\ RXU PRGHO )LJ ' VKRZHG D VHQRO\WLF HIIHFW IXOO\ FRPSDUDEOH WR WKH SRVLWLYH FRQWURO RXDEDLQ )LJ % DQG)LJ' :H SHUIRUPHG D VHFRQG YDOLGDWLRQ RI WKH HIIHFWLYHQHVV RI RXU PDFKLQH OHDUQLQJ PRGHOV XVLQJ D GLIIHUHQW VWUHVVRU :H IRFXVVHG RQ WKHUDS\LQGXFHG VHQHVFHQFH 7,6 6XSSOHPHQWDU\ )LJ $ ZKHUH KXPDQ HSLWKHOLDO FDQFHURXV FHOOV $ ZHUH LQGXFHG WR EHFRPH VHQHVFHQW E\ DGGLWLRQ RI HWRSRVLGH >@ &HOOV ZHUH WUHDWHG ZLWK  ȝ0 HWRSRVLGH IRU  KRXUV IROORZHG E\ DQRWKHU WKUHH GD\V RI H[SRVXUH WR PHGLD LQ VWDQGDUG FRQGLWLRQV $V D SRVLWLYH FRQWURO ZH HPSOR\HG QDYLWRFOD[ ,& FRQWURO  ȝȂ ,& VHQHVFHQFH  Q0 D %FO IDPLO\ LQKLELWRU ZLWK ZHOO FKDUDFWHULVHG VHQRO\WLF DFWLYLW\ >@ $GGLWLRQ RI  ȝ0 QDYLWRFOD[ WR $ FHOOV NLOOHG PRVW RI WKH FXOWXUHG VHQHVFHQW FHOOV EXW UHVXOWHG LQ QR UHGXFWLRQ LQ QXPEHUV RI QRQVHQHVFHQW FRQWURO FHOOV FRQILUPLQJ LWV RSWLPDO VHQRO\WLF DFWLYLW\6XSSOHPHQWDU\)LJ% &HOOV ZHUH WKHQ WUHDWHG ZLWK WKH WRS  KLWV LQ )LJ ' IURP RXU FRPSXWDWLRQDO VFUHHQ 6XSSOHPHQWDU\ )LJ $' 6XSSOHPHQWDU\ )LJ $ ZH IRXQG WKDW WKH VDPH WKUHH FRPSRXQGV YDOLGDWHG LQ RQFRJHQHLQGXFHG VHQHVFHQW FHOOV SHULSORFLQ ROHDQGULQ JLQNJHWLQ DOVR GLVSOD\HG VWURQJ VHQRO\WLF DFWLRQ LQ VHQHVFHQW $ FHOOV 7KH WKUHH FRPSRXQGV VKRZHG HQKDQFHG WR[LFLW\ ZKHQ FRPSDUHG ZLWK SUROLIHUDWLQJ QRQVHQHVFHQW $ FRQWUROV ZLWK DQ HIIHFW FRPSDUDEOH WR WKH SRVLWLYH FRQWURO +RHFKVW ODEHOOLQJ VKRZHG WKDW D VHQRO\WLF HIIHFW ZDV UHDFKHG ZLWK FRQFHQWUDWLRQV RI  Q0 ROHDQGULQ ,& FRQWURO  Q0 ,& VHQHVFHQFH  Q0  Q0 SHULSORFLQ ,& FRQWURO  Q0 ,& VHQHVFHQFH  Q0 DQG  ȝ0 JLQNJHWLQ ,& FRQWURO  ȝȂ ,& VHQHVFHQFH  ȝȂ 7KHVH GRVHV KDYH D PDUJLQDO HIIHFW RQ QRUPDO FHOOV EXW D GHFUHDVH LQ VXUYLYDO UDWH LQ VHQHVFHQW FHOOV 6XSSOHPHQWDU\ )LJ % 6XSSOHPHQWDU\ )LJ &' 7KH GRVHUHVSRQVH FXUYHV VKRZHG D ORZHUHG ,& IRU WKH WKUHH FRPSRXQGV LQ VHQHVFHQW FHOOV DV FRPSDUHG WR QRUPDO FHOOV 6XSSOHPHQWDU\ )LJ ' .CC-BY-NC-ND 4.0 International licensemade available 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 The copyright holder for this preprintthis version posted April 27, 2022. ; https://doi.org/10.1101/2022.04.26.489505doi: bioRxiv preprint 7R DVVHVV WKH VLPLODULW\ EHWZHHQ WKH WKUHH FRPSRXQGV FRUUHFWO\ LGHQWLILHG E\ WKH ;*%RRVW PRGHO )LJ ' DQG WKH VHQRO\WLFV HPSOR\HG IRU WUDLQLQJ )LJ % ZH FRPSXWHG WKH 7DQLPRWR GLVWDQFH LQ WKH GHVFULSWRU VSDFH EHWZHHQ JLQNJHWLQ ROHDQGULQ DQG SHULSORFLQ DQG HDFK RI WKH  VHQRO\WLFV LQ WKH WUDLQLQJ GDWD )LJ ( 0RUH WKDQ KDOI RI WKH WUDLQLQJ FRPSRXQGV ZHUH IRXQG WR EH PD[LPDOO\ GLVWDQW IURP RXU QHZO\ GLVFRYHUHG VHQRO\WLFV ZKLFK SURYLGHV VWURQJ YDOLGDWLRQ WKDW RXU PDFKLQH OHDUQLQJ DSSURDFKFDQHIIHFWLYHO\LGHQWLI\GLYHUVHFRPSRXQGVIRUVSHFLILFELRORJLFDOHIIHFWVVXFKDVVHQRO\VLV 'LVFXVVLRQ &XUUHQW DSSURDFKHV WR GUXJ GLVFRYHU\ VXIIHU IURP QRWRULRXVO\ KLJK DWWULWLRQ UDWHV LQ ODWHVWDJH SUHFOLQLFDO DQG FOLQLFDO GHYHORSPHQW 'XH WR WKHLU DELOLW\ WR SDUVH DQG GHWHFW SDWWHUQV LQ ODUJH YROXPHV RI GDWD $, KDV IRXQG DSSOLFDWLRQV DFURVV HYHU\ VWDJH RI WKH GUXJ GLVFRYHU\ SLSHOLQH >@ ,Q WKLV SDSHU ZH GHVFULEHG D VXFFHVVIXO PDFKLQH OHDUQLQJ DSSURDFK GHVLJQHG WR LGHQWLI\ QRYHO GUXJ FDQGLGDWHV LQ HDUO\ SKDVHV RI WKH GLVFRYHU\ SURFHVV :H IRFXVHG RQ WDUJHWHG HOLPLQDWLRQ RI VHQHVFHQW FHOOV D SKHQRW\SH WKDW KDV DWWUDFWHG VXEVWDQWLDO LQWHUHVW IRU DGMXYDQW FDQFHU WKHUDS\ >@ EXW IRU ZKLFK IHZ PROHFXODU WDUJHWV KDYH EHHQ LGHQWLILHG 2XU VWUDWHJ\ UHYHDOHG WKUHH FRPSRXQGV JLQNJHWLQ ROHDQGULQ SHULSORFLQ WKDW VHOHFWLYHO\ HOLPLQDWH FHOOV GLVSOD\LQJ RQFRJHQH DQG WKHUDS\LQGXFHG VHQHVFHQFH :H VKRZHG WKDW WKHVH FRPSRXQGV KDYH D SRWHQF\ FRPSDUDEOH WR VHQRO\WLFV SUHYLRXVO\ GHVFULEHG LQ WKH OLWHUDWXUH DQG FUXFLDOO\ RXU PHWKRG OHG WR ODUJH JDLQV LQ HIILFLHQF\ E\ UHGXFLQJ WKH QXPEHU RI FRPSRXQGVIRUH[SHULPHQWDOVFUHHQLQJE\PRUHWKDQIROG 2XU DSSURDFK RIIHUV VHYHUDO LQQRYDWLRQV WKDW GHSDUW IURP FXUUHQW SUDFWLFH LQ $, IRU GUXJ GLVFRYHU\ )LUVW LW UHOLHV VROHO\ RQ SXEOLVKHG GDWD IRU PRGHO WUDLQLQJ DQG WKXV DYRLGV WKH H[WUD FRVWV IRU LQKRXVH H[SHULPHQWDO FKDUDFWHULVDWLRQ RI WUDLQLQJ FRPSRXQGV 6HFRQG RXU PDFKLQH OHDUQLQJ PRGHOV ZHUH WUDLQHG RQ MXVW  FKHPLFDO VWUXFWXUHV ZLWK SURYHQ VHQRO\WLF DFWLRQ ZKLFK LV PXFK VPDOOHU GDWD WKDQ W\SLFDOO\ FRQVLGHUHG LQ WKH ILHOG WKH VL]H RI RXU WUDLQLQJ VHW LV D GLUHFW FRQVHTXHQFH RI VHQRO\VLV EHLQJ D UDUH PROHFXODU SURSHUW\ DQG WKH OLPLWHG QXPEHU RI VHQRO\WLFV UHSRUWHG LQ WKH OLWHUDWXUH VR IDU 7KH VXFFHVV RI RXU DSSURDFK GHPRQVWUDWHV WKDW PDFKLQH OHDUQLQJ FDQ WDNH PD[LPXP DGYDQWDJH RI OLWHUDWXUH GDWD HYHQ ZKHQ VXFK GDWD LV KHWHURJHQHRXV DQG RI PXFK VPDOOHU VFDOH WKDQ W\SLFDOO\ H[SHFWHG >@ 7KLUG RXU PRGHOV ZHUH WUDLQHG LQ D WDUJHWDJQRVWLF PDQQHU XVLQJ SKHQRW\SLF VLJQDWXUHV RI GUXJ DFWLRQ 7DUJHW VSHFLILFLW\ LV RI NH\ LPSRUWDQFH IRU GUXJ HIILFDF\ DQG VDIHW\ LQ ODWHU VWDJHV RI WKH GLVFRYHU\ SLSHOLQH EXW WKHUH DUH QXPHURXV FRQGLWLRQV RI KLJK HFRQRPLF DQG VRFLHWDO EXUGHQ ZLWK IHZ RU QR NQRZQ WDUJHWV >@ IRU VXFK FRQGLWLRQV WKHUH LV DQ RSSRUWXQLW\ IRU SKHQRW\SLF GUXJ GLVFRYHU\ WR LQFUHDVH WKH QXPEHU RI FKHPLFDO VWDUWLQJ SRLQWV WKDW FDQ EH FDUULHG WKURXJK WKH GLVFRYHU\SLSHOLQH >@ $ NH\ FKDOOHQJH LQ FRPSXWDWLRQDO GUXJ VFUHHQLQJ LV WKH FRQVWUXFWLRQ RI QXPHULFDO UHSUHVHQWDWLRQV RI FKHPLFDO VWUXFWXUHV WKDW DUH SUHGLFWLYH RI GUXJ HIILFDF\ >@ :LWK WKH DGYHQW RI GHHS OHDUQLQJ DV WKH OHDGLQJ SDUDGLJP LQ WKH ILHOG >@ PDQ\ UHFHQW ZRUNV KDYH GHYHORSHG VXFK UHSUHVHQWDWLRQV ZLWK HJ WUDQVIRUPHU PRGHOV IRU SUHGLFWLRQ RI FKHPLFDO UHDFWLRQV >@ JUDSK QHXUDO QHWZRUNV WR GHVFULEH PROHFXODU VWUXFWXUHV >@ DQG JHQHUDWLYH PRGHOV IRU GH QRYR FRPSRXQG GHVLJQ >@ ,Q RXU DSSURDFK ZH IRXQG WKDW FODVVLF SK\VLFRFKHPLFDO GHVFULSWRUV >@ ZHUH VXIILFLHQW WR WUDLQ XVHIXO PRGHOV DQG PRUHRYHU ZH REVHUYHG OLPLWHG EHQHILWV LQ WKH XVH RI GHHS OHDUQLQJ IRU FRPSRXQG IHDWXULVDWLRQ 6XSSOHPHQWDU\ 7DEOH  SRVVLEO\ DV D UHVXOW RI WKH VPDOO VL]H RI RXU WUDLQLQJ GDWD :H LQVWHDG IRXQG WKDW FDUHIXO GDWD DVVHPEO\ FXUDWLRQ DQG TXDOLW\ FRQWURO ZHUH WKH NH\ IRU VXFFHVV 6LQFH QHJDWLYH DVVD\V DUH UDUHO\ UHSRUWHG LQ WKH OLWHUDWXUH ZH EXLOW WKH WUDLQLQJ GDWD E\ SDLULQJ WKH NQRZQ VHQRO\WLFV ZLWK D EDFNJURXQG RI FRPSRXQGV DVVXPHG WR ODFN VHQRO\WLF DFWLRQ EXW ZLWK DQ DSSURSULDWH FKHPLFDO GLYHUVLW\ DQG D VL]H GHOLEHUDWHO\ FKRVHQ WR UHIOHFW WKH SDXFLW\ RI VHQRO\WLF FRPSRXQGV 7KHVH GHVLJQ FKRLFHV FDXVHG D KHDY\ LPEDODQFH EHWZHHQ WKH QXPEHU RI VHQRO\WLF DQG QRQVHQRO\WLF .CC-BY-NC-ND 4.0 International licensemade available 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 The copyright holder for this preprintthis version posted April 27, 2022. ; https://doi.org/10.1101/2022.04.26.489505doi: bioRxiv preprint FRPSRXQGV ZKLFK LQWURGXFHG DGGLWLRQDO FKDOOHQJHV IRU PRGHO WUDLQLQJ 6HYHUDO FKHFNV ZHUH QHHGHG WR HQVXUH WKDW WKH WUDLQLQJ GDWD ZDV GLYHUVH HQRXJK DQG DYRLGHG ELDV WRZDUG VSHFLILF FKHPLFDO FODVVHV 0RUHRYHU RXU PRGHOV JHQHUDOO\ GLVSOD\HG SRRU SHUIRUPDQFH DV TXDQWLILHG E\ FRPPRQ FODVVLILFDWLRQ PHWULFV SURGXFLQJ ODUJH QXPEHUV RI IDOVH SRVLWLYHV DQG IDOVH QHJDWLYHV LQ RXU FURVVYDOLGDWLRQ DQDO\VHV :H PLWLJDWHG WKH LPSDFW RI FODVV LPEDODQFH E\ SULRULWLVLQJ PRGHOV ZLWK D ORZHU QXPEHU RI IDOVH SRVLWLYHV VR DV WR UHGXFH WKH GRZQVWUHDP FRVWV IRU H[SHULPHQWDO YDOLGDWLRQ 2XU UHVXOWV WKXV VKRZ WKDW VHHPLQJO\ SRRU PRGHOV FDQ VWLOO EH HIIHFWLYH ZLWK DGHTXDWH FKHFNV DQG EDODQFHV RQ WKH VWUXFWXUH RI WKH WUDLQLQJ GDWD SOXV D FDUHIXO LQWHUSUHWDWLRQ RI SHUIRUPDQFH PHWULFV DQG PLVFODVVLILFDWLRQHUURUV 2XU DSSURDFK OHG WR D VLJQLILFDQW UHGXFWLRQ LQ H[SHULPHQWDO VFUHHQLQJ FRVWV ODUJHO\ EHFDXVH DOO PRGHOV ZHUH WUDLQHG VROHO\ RQ SXEOLVKHG GDWD DQG XQOLNH RWKHU UHFHQW VXFFHVVHV LQ WKH ILHOG >@ WKHUH ZDV QR QHHG WR VFUHHQ FRPSRXQGV SXUSRVHO\ IRU PRGHO WUDLQLQJ 7KH DSSURDFK WKXV RIIHUV H[FLWLQJ SURVSHFWV IRU QHZ RSHQ VFLHQFH DSSURDFKHV WR GUXJ GLVFRYHU\ 0RVW UHFHQWO\ WKH &29,' SDQGHPLF VSXUUHG D PXOWLWXGH RI VXFK LQLWLDWLYHV DFURVV WKH JOREH ZLWK WKH JRDO RI ILQGLQJ QHZ DQWLYLUDOV IURP WKH WURYHV RI SXEOLVKHG GDWD >@ 2XU ZRUN SURYLGHV D FRQFUHWH H[DPSOH RI D VLPSOH \HW HIIHFWLYH PDFKLQH OHDUQLQJ SLSHOLQH WKDW FDQ EH UHDGLO\ EXLOW IURP SXEOLVKHG VFUHHQLQJ GDWD :H KRSH WKLV DSSURDFK ZLOO FDWDO\VH PRUH RSHQ VFLHQFH DSSURDFKHV WR GLVFRYHU WUHDWPHQWV IRU FRQGLWLRQV RI XQPHW QHHG SDUWLFXODUO\ WKRVH IRU ZKLFK WKHUH LV D OLPLWHG JUDVS RI WKH ELRORJLFDO SDWKZD\V LQYROYHG LQGLVHDVHRQVHWDQGSURJUHVVLRQ 0HWKRGVDQG0DWHULDOV $'DWDDVVHPEO\IHDWXULVDWLRQDQGTXDOLW\FRQWURO 7UDLQLQJ GDWD :H DVVHPEOHG D OLVW RI  SUHYLRXVO\ LGHQWLILHG VHQRO\WLFV PLQHG IURP  VRXUFHV >±±@ )LJ % 7KH OLEUDU\ RI QHJDWLYH FRPSRXQGV FRQWDLQV  FRPSRXQGV IURP WKH /23$& /LEUDU\ RI 3KDUPDFRORJLFDOO\ $FWLYH &RPSRXQGV 0HUFN 'DUPVWDGW *HUPDQ\ DQG 3UHVWZLFN )'$DSSURYHG 3UHVWZLFN &KHPLFDO ,OONLUFK )UDQFH :H UHDVRQHG WKDW WKHVH OLEUDULHV DUH VXIILFLHQWO\ GLYHUVH IRU WUDLQLQJ PDFKLQH OHDUQLQJ PRGHOV $OWKRXJK DOO WKH FRPSRXQGV IURP WKHVH WZR OLEUDU\ VRXUFHV ZHUH GHHPHG DV QHJDWLYH WKHUH LV D SRVVLELOLW\ WKDW VRPH RI WKHVH PROHFXOHV KDYH EHHQ LQFRUUHFWO\ ODEHOOHG 7KLV LV EHFDXVH QRW DOO VHQRO\WLFV IRXQG KDYH EHHQ H[SUHVVO\ QDPHG LQ DOO KLJKWKURXJKSXW SXEOLFDWLRQV VRPH VRXUFHV RQO\ QDPH D VPDOO VHW RI WKHLU GLVFRYHULHV RU EHFDXVH WKHVH PROHFXOHV KDYH RQO\ EHHQ WHVWHG LQ RQH RU WZR FHOO OLQHV XQGHU RQH W\SHRIVHQRO\WLFLQGXFWLRQSURFHVV )HDWXULVDWLRQ 7KH WUDLQLQJ GDWDVHW FRQWDLQV D RQHGLPHQVLRQDO YHFWRULDO UHSUHVHQWDWLRQ RI WKH WZRGLPHQVLRQDO VWUXFWXUH RI HYHU\ PROHFXOH LQ WKH IRUP RI D 60,/(6 VWULQJ 7KH PDMRULW\ RI 60,/(6 ZHUH WDNHQ IURP WKH OLEUDU\ RI RULJLQ RI HYHU\ FRPSRXQG /23$& RU 3UHVWZLFN IRU WUDLQLQJ 6HOOHFN RU 7DUJHW0RO IRU VFUHHQLQJ ZLWK WKH H[FHSWLRQ RI ILYH SRVLWLYHV 3UR'UXJ $ -+%% &*3$ $ $ ZKLFK 60,/(6 ZHUH FDOFXODWHG XVLQJ &KHP'UDZ >@ )RU FKLUDO PROHFXOHV ZH IDYRXUHG LVRPHULF 60,/(6 UHSUHVHQWDWLRQV LQVWHDG RI WKH FDQRQLFDO FDVH :H HPSOR\HG WKH 5'.LW SDFNDJH >@ WR FRPSXWH  SK\VLFRFKHPLFDO GHVFULSWLRQV IRU HDFK FRPSRXQG ZKLFKTXDQWLI\GLIIHUHQWDVSHFWVRIWKHPROHFXODUVWUXFWXUHLWVIUDJPHQWVDQGJOREDOFKDUDFWHULVWLFV &OXVWHULQJ DQDO\VLV DQG 7DQLPRWR JUDSK 7R TXDQWLI\ WKH GLYHUVLW\ RI WKH VHQRO\WLFV HPSOR\HG IRU WUDLQLQJ )LJ &( ZH SHUIRUPHG NPHDQV FOXVWHULQJ RI WKH  SRVLWLYHV XVLQJ WKH 5'.LW GHVFULSWRUV DV IHDWXUH YHFWRUV DQG FRVLQH GLVWDQFH IXQFWLRQ EHWZHHQ ]VFRUH QRUPDOLVHG IHDWXUH YHFWRUV 7KH GHJUHH RI FOXVWHULQJ ZDV TXDQWLILHG E\ WKH NPHDQV VFRUH )LJ & LH WKH RSWLPDO YDOXH RI WKH ORVV IXQFWLRQ HPSOR\HG WR GHWHUPLQH WKH FOXVWHULQJ 7KH TXDOLW\ RI WKH NPHDQV FOXVWHUV ZHUH GHWHUPLQHG .CC-BY-NC-ND 4.0 International licensemade available 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 The copyright holder for this preprintthis version posted April 27, 2022. ; https://doi.org/10.1101/2022.04.26.489505doi: bioRxiv preprint ZLWK WKH VLOKRXHWWH FRHIILFLHQW 6DYHUDJHG DFURVV WKH  VHQRO\WLFV 7KH VLOKRXHWWH FRHIILFLHQW YDULHV EHWZHHQ  DQG  ZLWK 6  LQGLFDWLQJ WKDW FRPSRXQGV DUH LQ ZHOO VHSDUDWHG FOXVWHUV 6  LQGLFDWLQJ RYHUODSSLQJ FOXVWHUV DQG 6  LQGLFDWLQJ LQFRUUHFW DVVLJQPHQW RI FOXVWHUV 7R EXLOG WKH 7DQLPRWR GLVWDQFH JUDSK )LJ ' ZH ILUVW FRQVWUXFWHG D IXOO\ FRQQHFWHG JUDSK ZHLJKWHG E\ WKH SDLUZLVH 7DQLPRWR GLVWDQFHV IHDWXUH YHFWRUV 7KLV JUDSK ZDV WKHQ VSDUVLILHG ZLWK WKH NQHDUHVW QHLJKERXUV JUDSK N  LQWHUVHFWHG ZLWK WKH PLQLPXP VSDQQLQJ WUHH RI WKH RULJLQDO JUDSK 7KH HGJH ZLGWKV ZHUH VHW DV WKH 7DQLPRWR VLPLODULW\ EHWZHHQ FRPSRXQGV GLVWDQFH &OXVWHULQJ RI WKH 7DQLPRWR GLVWDQFH JUDSK )LJ ( ZDV GRQH ZLWK D 0DWODE LPSOHPHQWDWLRQ RI WKH /RXYDLQ DOJRULWKP IRU FRPPXQLW\ GHWHFWLRQ >@ 7R FRPSDUH WKH /RXYDLQ FOXVWHUV ZLWK WKH FRPSRXQGV ODEHOOHG DFFRUGLQJ WR WKHLU VRXUFH )LJ % ZH HPSOR\HG WKH DGMXVWHG 5DQG ,QGH[ ZKLFK LV D PHDVXUH RI WKH VLPLODULW\ EHWZHHQ WZR FOXVWHULQJV DGMXVWHG IRU WKH FKDQFH JURXSLQJ RI FRPSRXQGV >@ ORZ YDOXHV RI WKH $5, LQGLFDWH OLWWOHVLPLODULW\EHWZHHQFOXVWHULQJV %0RGHOWUDLQLQJDQGFRPSXWDWLRQDOVFUHHQ $OO PRGHOV ZHUH WUDLQHG ZLWK WKH VFLNLWOHDUQ  3\WKRQ OLEUDU\ SOXV WKH ;*%RRVW Y  >@ 0RGHOV ZHUH WUDLQHG RQ D UHGXFHG VHW RI  ]VFRUH QRUPDOLVHG IHDWXUHV LGHQWLILHG DV UHOHYDQW IRU FODVVLILFDWLRQ XVLQJ VFLNLWOHDUQ IHDWXUH LPSRUWDQFH ZLWK D IRUHVW RI WUHHV IXQFWLRQ 6XSSOHPHQWDU\ )LJ  $OOPRGHOVZHUHVFRUHGZLWKWKUHHPHWULFVRIFODVVLILFDWLRQ SHUIRUPDQFH DFFXUDF\ 73717371)3)1 SUHFLVLRQ 7373)3 UHFDOO 7373)1 ZKHUH 73 71 )3 DQG )1 DUH WKH QXPEHU RI WUXH SRVLWLYHV WUXH QHJDWLYHV IDOVH SRVLWLYHV DQG IDOVH QHJDWLYHV UHVSHFWLYHO\ 0RGHOV ZHUH WUDLQHG DQG HYDOXDWHG ZLWK D UDQGRP  VSOLW RI WKH GDWD  IRU WUDLQLQJ DQG  IRU WHVWLQJ :H VWUDWLILHG VDPSOHV LQ WKH UDQGRP VSOLW WR HQVXUH WKDW ERWK WKH SRVLWLYH DQG QHJDWLYH FODVVHV ZHUH SURSRUWLRQDOO\ UHSUHVHQWHG LQ WKH WUDLQLQJ DQG WHVWLQJ VHWV ,Q DOO FDVHV D IROG FURVVYDOLGDWLRQ ZDV SHUIRUPHG )RU WKH ILQDO ;*%RRVW PRGHO HPSOR\HG IRU WKH FRPSXWDWLRQDO VFUHHQ ZH RSWLPLVHG WKH K\SHUSDUDPHWHU WUHHGHSWK DQG REWDLQHG FRQVLVWHQW SHUIRUPDQFH RYHU IROG FURVVYDOLGDWLRQ UXQV ZLWK YDOXHV YDULHG IURP  WR  ZH VHWWOHG WUHHGHSWK  WR DOORZ IRU D PRUH SURIRXQG XVH RI RXU KLJKGLPHQVLRQDO IHDWXUH VSDFH )RU WKH FRPSXWDWLRQDO VFUHHQ ZH XVHG WKH / 7DUJHW0RO $QWLFDQFHU 7DUJHW0RO &KHPLFDOV :HOOHVOH\ +LOOV 0$ DQG / 6HOOHFN )'$DSSURYHG  3DVVHG 3KDVH 6HOOHFN &KHPLFDOV +RXVWRQ 7; OLEUDULHV GXH WR WKH YHUVDWLOLW\ RI FRPSRXQGV UHSUHVHQWHG LQ WKHP :H XVHG WKH SUHWUDLQHG ;*%RRVW PRGHO WR FRPSXWH SUHGLFWLRQ VFRUHV RQ WKH DXJPHQWHG OLEUDU\ LH WKH SUREDELOLW\ RI D FRPSRXQG EHLQJ FODVVLILHG DV VHQRO\WLF)LJ& &9DOLGDWLRQDVVD\V :H SHUIRUPHG H[SHULPHQWDO YDOLGDWLRQ RI WKH FRPSRXQGV ZLWK ]VFRUH! )LJ ' 7KHVH FRUUHVSRQG WR  FRPSRXQGV RXW RI D WRWDO RI  FRPSRXQGV :H HPSOR\HG WZR FHOOXODU PRGHOV RI VHQHVFHQFH RQH RI 2,6 DQG D VHFRQG RQH RI 7,6 )RU WKH 2,6 FDVH ZH XWLOLVHG ,05 (55$6 FHOOV ZLWK 2+7 DW  Q0 7KH 2+7 WUHDWPHQW KDG D GXUDWLRQ RI VL[ GD\V )RU WKH 7,6 PRGHO ZH XVHG $ FHOOV ZLWK HWRSRVLGH DW  ȝ0 7KH H[SRVXUH RI WKH FHOOV WR HWRSRVLGH ODVWHG  KRXUV DIWHU ZKLFKSHULRGWKHFHOOVZHUHFXOWLYDWHGIRUDIXUWKHUWKUHHGD\VZLWKQRUPDOPHGLD &HOO FXOWXUH ,05 IHPDOH KXPDQ IRHWDO OXQJ ILEUREODVWV DQG $ KXPDQ OXQJ DGHQRFDUFLQRPD FHOOV ZHUH REWDLQHG IURP WKH $PHULFDQ 7\SH &XOWXUH &ROOHFWLRQ $7&& 0DQDVVDV 9$ 7KH FHOO OLQHV ZHUH FRQILUPHG WR EH P\FRSODVPD QHJDWLYH /RQ]D 0\FR$OHUW FDW /7 ,05 (55$6 LV D .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 27, 2022. ; https://doi.org/10.1101/2022.04.26.489505doi: bioRxiv preprint )LJXUH  7UDLQLQJ RI PDFKLQH OHDUQLQJ PRGHOV DQG FRPSXWDWLRQDO VFUHHQLQJ $ 3LSHOLQH IRU PRGHO WUDLQLQJ FRPSRXQG VFUHHQLQJ DQG KLW YDOLGDWLRQ 6HYHUDO FODVVLILFDWLRQ VFRUHV ZHUH XVHG DV SHUIRUPDQFH PHWULFV WR GHWHUPLQH WKH PRVW VXLWDEOH PRGHO IRU WKH FRPSXWDWLRQDO VFUHHQ % 5HVXOWV IURP WKUHH PDFKLQH OHDUQLQJ PRGHOV WUDLQHG RQ  FRPSRXQGV )LJXUH $ EDU SORWV VKRZ DYHUDJH SHUIRUPDQFH PHWULFV FRPSXWHG LQ IROG FURVV YDOLGDWLRQ ZLWK HUURU EDUV GHQRWLQJ RQH VWDQGDUG GHYLDWLRQ DFURVV IROGV 7KH FRQIXVLRQ PDWULFHV ZHUH FRPSXWHG IURP PRGHOV WUDLQHG RQ  RI FRPSRXQGV DQG WHVWHG RQ  SRVLWLYHV DQG  QHJDWLYHV WKDW ZHUH KHOGRXW IURP WUDLQLQJ $OO PRGHOV GLVSOD\HG SRRU SHUIRUPDQFH PHWULFV 6XSSOHPHQWDU\ 7DEOH  DQG ZH FKRVH WKH ;*%RRVW DOJRULWKP IRU VFUHHQLQJ EHFDXVH RI LWV ORZHU QXPEHU RI IDOVH SRVLWLYHV WKH KLJK DFFXUDF\ RI DOO PRGHOV LV D ZHOONQRZQ DUWHIDFW LQ FODVVLILFDWLRQ SUREOHPV ZLWK KHDY\ FODVV LPEDODQFH & 5HVXOWV IURP FRPSXWDWLRQDO VFUHHQ RI WKH / 7DUJHW0RO $QWLFDQFHU DQG / 6HOOHFN )'$DSSURYHG  3DVVHG 3KDVH FKHPLFDO OLEUDULHV WRWDOLQJ  FRPSRXQGV 7KH ;*%RRVW PRGHO LV KLJKO\ VHOHFWLYH DQG VFRUHV WKH YDVW PDMRULW\ RI FRPSRXQGV ZLWK DQ H[WUHPHO\ ORZ SUREDELOLW\ RI KDYLQJ VHQRO\WLF DFWLRQ D VPDOO IUDFWLRQ RI 1  FRPSRXQGV ZHUH VFRUHG ZLWK 3! ZKLFK ZH VHOHFWHG IRU H[SHULPHQWDO YDOLGDWLRQ ' &RPSRXQGV VHOHFWHG IRU VFUHHQLQJ UDQNHG DFFRUGLQJ WR WKHLU ]VFRUH QRUPDOLVHG SUHGLFWLRQ VFRUHV IURP WKH ;*%RRVW PRGHO WKH VHOHFWHG FRPSRXQGV DUH IDU RXWOLHUV LQ WKH GLVWULEXWLRQ RI SDQHO & ( 7ZR GLPHQVLRQDO W61( YLVXDOLVDWLRQ RI DOO FRPSRXQGV HPSOR\HG LQ WKLV ZRUN W61( SORWV ZHUH JHQHUDWHG ZLWK SHUSOH[LW\  OHDUQLQJ UDWH  DQG PD[LPDO QXPEHU RI LWHUDWLRQV  >@ 3UHGLFWLRQV VFRUHV DERYHIURPWKH;*%RRVWPRGHODUHPDUNHGZLWKRUDQJHFLUFOHV .CC-BY-NC-ND 4.0 International licensemade available 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 The copyright holder for this preprintthis version posted April 27, 2022. ; https://doi.org/10.1101/2022.04.26.489505doi: bioRxiv preprint )LJXUH  ([SHULPHQWDO FKDUDFWHULVDWLRQ RI FRPSRXQGV VHOHFWHG IRU VFUHHQLQJ LQ RQFRJHQHLQGXFHG VHQHVFHQW FHOOV $ ([SHULPHQWDO VHWXS RI RQFRJHQHLQGXFHG VHQHVFHQFH PRGHO ZLWK ,05 (55$6 FHOOV 6HQHVFHQFH ZDV LQGXFHG E\ DGGLWLRQ RI 2+7 DW  Q0 GXULQJ WKHW GXUDWLRQ RI WKH H[SHULPHQW HLJKW GD\V &RQWURO DQG VHQHVFHQW FHOOV ZHUH SODWHG LQ D ZHOO SODWH GD\ ILYH RI 2+7 LQGXFWLRQ 7RS SUHGLFWHG FRPSRXQGV ZHUH DGGHG DIWHU PXOWLZHOO VHHGLQJ DQG  KRXUV DIWHUZDUGV WKH FHOOV ZHUH IL[HG DQG WKH QXFOHL VWDLQHG DQG FRXQWHG % %DU SORW RI 2,6 SRVLWLYH H[SHULPHQWDO FRQWURO RXDEDLQ DW  Q0 'DWD LV QRUPDOLVHG WR '062 0HDQ  VG DUH VKRZQ IURP Q  H[SHULPHQWV & 5HVXOWV IURP H[SHULPHQWDO YDOLGDWLRQ RI FRQWUROV DQG WKH WRS  FRPSRXQGV IURP )LJ ' SUHGLFWHG WR KDYH VHQRO\WLF DFWLRQ ZLWK 3! 7KUHH FRPSRXQGV RXW RI WKH  GLVSOD\HG VHQRO\WLF DFWLYLW\ JLQNJHWLQ ROHDQGULQ DQG SHULSORFLQ ' 'RVHUHVSRQVH FXUYHV RI WKH WKUHH QHZO\ IRXQG VHQRO\WLF FRPSRXQGV 7KH VHQRO\WLF LQGH[ 6, LV GHILQHG DV WKH UDWLR EHWZHHQ WKH ,& RI FRQWURO FHOOV DQG WKH ,& RI VHQHVFHQW FHOOV 'DWD LV QRUPDOLVHG WR '062 0HDQ  VG DUH VKRZQ IURP Q  H[SHULPHQWV ( 7DQLPRWR GLVWDQFH IURP HDFK RI WKH WKUHH YDOLGDWHG VHQRO\WLFV WR HDFK VHQRO\WLF HPSOR\HG IRU WUDLQLQJ GLVWDQFHV ZHUH FDOFXODWHG XVLQJ WKH 5'.LW GHVFULSWRUV WKDW ZHUH HPSOR\HG LQ WKH WUDLQLQJ RI PDFKLQH OHDUQLQJ PRGHOV LQ )LJ % DQG 6XSSOHPHQWDU\7DEOH .CC-BY-NC-ND 4.0 International licensemade available 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 The copyright holder for this preprintthis version posted April 27, 2022. ; https://doi.org/10.1101/2022.04.26.489505doi: bioRxiv preprint