>> GOOD AFTERNOON, EVERYONE. IT'S A GREAT PLEASURE TO WELCOME YOU TO THE FIRST WEDNESDAY AFTERNOON LECTURE OF THIS PARTICULAR ACADEMIC SEASON. AND IT'S GREAT TO SEE SO MANY PEOPLE HERE IN MASUR, AND I KNOW THERE ARE MANY OTHERS WATCHING BY VIDEO, AND WE WELCOME YOU AS WELL. WE DO THIS ON MOST WEDNESDAYS, WE BRING IN AN AMAZING SET OF LECTURERS THAT COME HERE TO TELL US THEIR LATEST RESULTS. NEXT WEEK, BY THE WAY, WILL BE CARL JEUNE, ONE OF THE LEADERS OF CANCER IMMUNOTHERAPY FROM THE UNIVERSITY OF PANCREATIC ADENOCARCINOMA, SO I'M SURE YOU WANT TO COME AND HEAR WIT ABOUT
THAT AS WELL. PERSONAL PLEASURE FOR ME TO BE ABLE TO INTRODUCE TODAY'S SPEAKER, DAVID REICH. HIS WORK IN THE AREA OF UNDERSTANDING HUMAN ORIGINS IS AT THE VERY LEADING EDGE ON QUESTIONS THAT ALL OF US HAVE ASKED FOR SENT KREN CENTURY,
MILLENNIUMS, ABOUT WHERE WE CAME FROM AND HOW WE'RE ALL RELATED, AND USING THE TOOLS OF MODERN GENOMICS AND SOME PRETTY GOOD SLEUTHING AND EPIDEMIOLOGY AND FIGURING OUT HOW TO DERIVE DNA FROM ANCIENT BONES. DAVID HAS BEEN SUCCESSIVELY, OVER THE SPACE OF SEVERAL YEARS, ADDING MORE AND MORE DETAILS TO THAT MAP OF INFORMATION ABOUT OUR ORIGINS AND HOW ALL OF US, ORIGINALLY AFRICANS, HAVE ENDED UP IN OTHER PARTS OF THE WORLD, AND HOW THE STUDY OF DNA CAN TEACH US ABOUT THOSE HISTORICAL EVENTS IN WAYS THAT HAVE REALLY REMARKABLE DETAIL ATTACHED TO THEM. IT'S ALSO A PARTICULAR PLEASURE TO HAVE DAVID HERE AS OUR SPEAKER BECAUSE IT'S KIND OF A FAMILY THING, MANY OF YOU KNOW DAVID'S BROTHER, DANNY REICH, WHO'S SITTING THERE IN THE BLUE SHIRT, ONE OF OUR DISTINGUISHED PRINCIPAL INVESTIGATORS IN THE NEUROIMAGING ARENA AND SOMEBODY THAT WE HAVE ASKED MANY TIMES TO TALK ABOUT HIS RESEARCH TO DISTINGUISHED VISITORS AT VARIOUS MOMENTS, INCLUDING HIGH-RANKING OFFICIALS OF THE UNITED STATES GOVERNMENT BUT WHO DOES WONDERFUL RESEARCH IN THE AREA OF MULTIPLE SCLEROSIS AND OTHER THINGS. AND I'M ALSO VERY HAPPY TO POINT OUT THAT DANNY AND DAVID'S PARENTS ARE HERE ALSO, SITTING ON EITHER SIDE OF HIM, AND HIS DAD WAS ONCE HERE AS NIH AS AN INVESTIGATOR IN NIMH. SO THIS IS REALLY A WONDERFUL FAMILY OCCASION THAT WE HAVE HERE. WELL, TO SAY VERY BRIEFLY A COUPLE WORDS ABOUT DAVID SO I DON'T TAKE ANY MORE OF HIS TIME, HE IS CURRENTLY PROFESSOR OF GENETICS AT HARVARD, ASSOCIATE MEMBER OF THE BROAD INSTITUTE, HE'S HAN INVESTIGATOR IN THE HOWARD HUGHES MEDICAL INSTITUTE, AND HIS UNDERGRADUATE WORK IN HARVARD AS PHYSICS, ALWAYS A GOOD START FOR SOMEBODY WHO'S GOING TO BE PRETTY DARN QUANTITATIVE AND PRETTY MATHEMATICAL AS WELL. AFTER THAT, STATISTICAL GENETICS AT THE UNIVERSITY OF OXFORD, WHERE HE WAS WITH DAVID GOLDSTEIN, AND THEN ON TO HARVARD MEDICAL SCHOOL AND MIT, BUT ENDING UP AT THE WHITEHEAD FOR A POSTDOC IN MEDICAL GENETICS THEN TO HARVARD AND THE APPOINTMENTS I JUST MENTIONED AT THE BRODE AND HOWARD HUGHES. HIS WORK HAS BEEN RECOGNIZED BY A VARIETY OF SPECIAL AWARDS, THE NEW CLEVELAND PRIZE FROM AAAS FOR THE BEST PAPER IN SCIENCE OF THE YEAR IN 2010, THE OUTSTANDING ACHIEVEMENT AWARD NAMED AS ONE OF NATURE'S 10 PEOPLE WHO MATTER THIS YEAR BY "NATURE" MAGAZINE LAST YEAR. HAPPY TO CLAIM HIM AS A P.I. AND CO-P.I. ON SEVERAL GRANTS FROM NIH, AND AS YOU WILL HEAR, HAPPY TO SAY THE WAY IN WHICH HIS RESEARCH IS EVOLVING IN FASCINATING WAYS, INCLUDING, BY THE WAY, THAT HIS PAPER IN "NATURE" CAME OFF EMBARGO AT 1:00 TODAY, SO IT'S A VERY GOOD BIT OF TIMING TO HAVE HIM HERE TO TALK ABOUT THE LATEST FINDINGS. SO PLEASE JOIN ME IN WELCOMING DR. DAVID REICH. [APPLAUSE] >> DO PEOPLE HEAR ME? THANK YOU. IT'S REALLY A GREAT HONOR TO COME SPEAK HERE AND TO BE INVITED TO GIVE THIS LECTURE TO THE NIH. I'M GOING TO TELL YOU ABOUT A NEW SCIENTIFIC INSTRUMENT THAT'S BEEN DEVELOPED IN THE LAST REALLY FIVE OR SIX YEARS AND THAT'S REALLY MADE IT POSSIBLE TO STUDY THINGS THAT WERE NOT POSSIBLE TO STUDY BEFORE, AND I'M GOING TO TELL YOU ABOUT WORK THAT I'VE BEEN INVOLVED IN, IN TRYING TO USE THIS SCIENTIFIC INSTRUMENT TO LEARN NEW THINGS. SO I HOPE THAT AT THE END OF THIS LECTURE, YOU WILL APPRECIATE THE POWER OF THIS NEW TOOL AND ITS POTENTIAL TO LEARN A LOT OF NEW THINGS. SO I'M FIRST GOING TO TELL YOU ABOUT THIS NEW SCIENTIFIC INSTRUMENT, ANCIENT DNA, SO THIS IS AN INSTRUMENT THAT WAS NOT INVENTED ON THIS CONTINENT, IT WAS REALLY DEVELOPED ALMOST ENTIRELY IN EUROPE BY EUROPEAN LABORATORIES, ESPECIALLY PAVOS LABORATORY IN GERMANY, BUT NOW HAS BECOME A TECHNOLOGY THAT OTHER GROUPS CAN USE AS WELL BECAUSE OF THE WORK DONE THERE AND ELSEWHERE. SO IT STARTS WITH A BONE FROM IN THIS CASE A HUMAN. I WORK ON HUMAN BONES MOSTLY, BUT OTHER LABORATORIES WORK ON OTHER TYPES OF SKELETAL REMAINS. WE TAKE THE BONE INTO A CLEAN ROOM, WHICH IS A CONTROLLED ENVIRONMENT WHERE THERE'S POSITIVE PRESSURE SO THAT THE DNA FROM THE OUTSIDE THE CLEAN ROOM DOESN'T GET INSIDE THE CLEAN ROOM. THE DANGER IS CONTAMINATION OF THESE SAMPLES WHICH HAVE SO LITTLE DNA IN THE FIRST PLACE. THERE IS UV LIGHT WHEN PEOPLE AREN'T THERE TO CROSS LINK THE DNA SO THAT YOU DON'T SEQUENCE DNA THAT'S FROM THE RESEARCHERS IN THE ROOM. PEOPLE WEAR BODY SUITS AND THERE'S A LOT OF CLEANING WITH BLEACH AND USE OF HOODS IN ORDER TO CLEAN SAMPLES. ONE DRILLS BENEATH THE SURFACE OF THE BONE WITH A LITTLE BORE HOLE TO REMOVE -- TO TRY TO GET BEYOND WHERE IT MIGHT HAVE BEEN CONTAMINATED BY ARCHAEOLOGISTS OR WHOA PEOPLE WHO HAVE HANDLED
THE BONE, IT CONTAINS A BIT OF POWDER. THAT POWDER IS REMINERALIZED AND PROTEINS ARE REMOVED THROUGH A SERIES OF CHEMICAL STEPS AND THEN TURNED INTO SEQUENCABLE FORM WITH ONE OF THE MODERN SEQUENCERS TO PRODUCE A LOT OF DATA. SO THAT'S THE BASIC SORT OF STEPSET OF STEPSUSED TO GET DNA
OUT OF SAM
PLES, AND THERE'S A WHOLE BUNCH OF OPTIMIZATION THAT HAS BEEN IMPLEMENTED IN ORDER TO MAKE THIS MORE EFFICIENT AND INCREASE THE AMOUNT OF DNA THAT ONE CAN GET OUT OF A SAMPLE BY LITERALLY 10,000 TIMES OVER THE LAST SIX YEARS. SO I'M JUST GOING TO START WITH A PICTURE OF MOORE'S LAW. SO MOORE'S LAW IS ABOUT COMPUTER WORLDS AND IT'S WHERE THE NUMBER OF ELEMENTS ON INTEGRATED CIRCUITS DOUBLES EVERY ONE TO TWO YEARS, AND THAT'S BEEN GOING ON FOR THE LAST CENTURY, AND IT'S DRIVEN THE COMPUTER EVOLUTION. SO THERE IS A MOORE'S LAW OF ANCIENT DNA BUT THE RISE HAS BEEN MUCH FASTER AND MORE RECENT, SO IN 2010 -- AND I'M TELLING YOU HERE ABOUT WHOLE GENOME DATA, NOT JUST MITOCHONDRIAL DNA SEQUENCES BUT JUST WHOLE GENOME DATA WHERE THERE'S ENOUGH TO REALLY LEARN ABOUT POPULATION VARIATION. SO IF YOU USE AS A CUTOFF AN AMOUNT WHERE YOU CAN BEGIN TO LEARN ABOUT POPULATION RELATIONSHIPS, THERE WERE FOUR SEQUENCES PUBLISHED IN 2010, THERE WERE A BURST OF 20 OR 30 SEQUENCES PUBLISHED IN 2014, AND THEN IT WENT INTO HYPERDRIVE IN 2015 WITH SEVERAL HUNDRED SEQUENCES AND IT'S INCREASING AND INCREASING, NOW WE HAVE MORE THAN A THOUSAND SEQUENCES IN OUR LABORATORY AND I'M SURE THAT'S TRUE ALSO OF OTHER LABORATORIES, AND THERE'S SO MUCH DATA BEING PRODUCED RIGHT NOW THAT THERE'S MUCH MORE DATA THAT'S UNPUBLISHED --
THAT'THANPUBLISHED JUST BEC
AUSE THERE'S NOT ENOUGH TIME TO PUBLISH IT. WHAT WE'VE DONE IN OUR LABORATORY, I GUESS IT'S A VERY AMERICAN THING DO, IS TO MAKE THESE BEAUTIFUL TECHNIQUES DEVELOPED IN EUROPE, BEAUTIFUL PIECE WORK TECHNIQUES, TO MAKE THEM INDUSTRIAL. AND SO WE TRY TO TAKE THE OLD APPROACH, WHICH IS STILL A WONDERFUL APPROACH, SEARCHING FOR A GOLDEN SAMPLE, SCREENING AND SCREENING AND SCREENING THESE SKELETAL REMAINS FOR SOME SAMPLE THAT WILL HAVE ENOUGH DNA THAT IT CAN BE SEQUENCED AND THEN WORKING VERY HARD ON IT, TO PROCESS WHERE WE SCREEN MANY, MANY SAMPLES AND USE -- AND TAKE ADVANTAGE OF EFFICIENCIES AND TECHNIQUES THAT MAKES IT POSSIBLE TO CONVERT A HIGHER FRACFRACTION OF SAMPLES TO
PROVIDE THE INFORMATION WE NEED. SO OUR GOAL WAS TO IMPLEMENT A SCREENING PROCESS THAT WOULD TURN A LARGE FRACTION OF THE SAMPLES WE ANALYZE INTO WORKING DATA, AND TO HAVE A REASONABLE COST PER SAMPLE TO PRODUCE GENOME-WIDE DATA, IN OUR CASE, LESS THAN $500 A SAMPLE. ONE OF THE REASONS ANCIENT DNA IS SO EXPENSIVE AND CONTINUES TO BE SO EXPENSIVE IN MOST IMPLEMENTATIONS IS THAT BECAUSE SO LITTLE ANCIENT DNA IS HUMAN, BONES ARE MOSTLY -- THE DNA YOU GET OUT OF BONES IS MOSTLY MICROBIAL, IT'S FROM THE BACTERIA AND FUNGI THAT ATE YOU WHEN YOU DIED, SO VERY LITTLE OF THE DNA THAT YOU SEQUENCE IS FROM THE INDIVIDUAL YOU WANT TO SEQUENCE. EVEN IF IT'S AN UNCONTAMINATED SAMPLE. BEYOND THAT, OF COURSE, MOST OF THE DNA THAT YOU SEQUENCE EVEN FROM A HUMAN IS NOT FROM THE SECTIONS OF THE GENOME YOU PROBABLY ARE GOING TO WANT TO ANALYZE, PERHAPS FOR STUDYING POPULATION RELATIONSHIPS. SO THE SOLUTION THAT WE TOOK WAS A SOLUTION WHICH IS AN ADAPTATION OF A METHOD DEVELOPED BY KIAOMEI FU AND MATHIAS MEYER, AND WHAT WE DO IS WE TAKE THE DNA SAMPLE, MADE INTO SEQUENCABLE FORM FROM THE DAB LIBRARY AND WE TAKE A TECHNIQUE, EXOME SEQUENCING, WE WASH IT OVER A SERIES OF 50 DNA LETTER LONG SEQUENCES THAT TARGET THE AREA OF THE GENOME WE'RE INTERESTED IN, AND WHAT'S LEFT IS NOT ONLY FROM HUMAN BUT THE REAMES WE'RE INTERESTED IN, SO MOST OF WHAT WE'RE LEFT WITH IS THE AREAS WE'RE INTERESTED IN AND THEN IT'S NOT A LOT OF SEQUENCING TO OBTAIN DATA FROM THESE APPROXIMATELY 1 MILLION POSITIONS IN THE GENOME. SO WE'RE CURRENTLY SCREENING AND PRODUCING -- SCREENING MORE THAN A THOUSAND SAMPLES PER YEAR IN OUR LABORATORY, WHICH IS A HUGE CHANGED COMPARED TO THE SITUATION IN THE PAST, WE'RE NOT THE ONLY LABORATORY THAT'S ABLE TO SCREEN SUBSTANTIAL NUMBERS OF SAMPLES BUT IT'S REALLY TRANSFORMATIVE IN TERMS OF WHAT IT ALLOWS US TO DO. SO THIS ARROWS U ALLOWS US TO
REDUCE COST AND IMPROVE QUALITY. THIS WAS A WONDERFUL EXAMPLE PUBLISHED LAST YEAR IN DENMARK WHICH TOOK ABOUT 100 SAMPLE, SEQUENCED THEM, BRUTE FORCE, BY RUNNING THE SAMPLES THROUGH THE SEQUENCER. THIS IS THE NUMBER OF SEQUENCES IN THE HUNDREDS OF THOUSANDS AND APPROXIMATELY THE SAME COST IN DOLLARS, AND THIS WAS A MILLION DOLLAR EXPERIMENT WHERE ON AVERAGE, ABOUT .20% OF POSITIONS IN THE GENOME WERE COVERED, BUT WHEN WE APPLY THIS APPROACH, WE HAVE MANY FEWER SEQUENCES REQUIRED AND THAT WE THINK ARE MOST INTERESTING TO ANALYZE. SO IT ALLOWS A GREAT AMOUNT OF EFFICIENCY AND MAKES IT ECONOMIC TALL TO DO THIS KIND OF WORK. I'M REQUESTING TO TELL YOU IN THE SECOND PART OF MY TALK A SPECIFIC EXAMPLE ABOUT HOW ANCIENT DNA HAS PROVIDED NEW INSIGHTS ABOUT THE PAST, AND I'M GOING TO TALK ABOUT EUROPE AND ABOUT WESTERN EURASIA MORE BROADLY, NOT BECAUSE THESE ARE PARTICULARLY IMPORTANT PLACES OF THE WORLD BUT IT'S BECAUSE WE KNOW THE MOST. SO THIS TECHNOLOGY OF ANCIENT DNA WAS DEVELOPED IN EUROPE, AND BECAUSE OF THAT WE HAVE MUCH MORE DATA, ABOUT 90% OF THE GENOME WIDE DATA IN THE LITERATURE IS FROM EUROPE OR WESTERN EUR ASIA, JUST AS INTERESTING THINGS CAN BE DONE IN THE AMERICAS, SOUTH ASIA, AFRICA, BUT I'M GOING TO TELL BUT EUROPE TO ILLUSTRATE THE POWER OF THESE APPROACHES. SO THE BACKGROUND FOR THIS IS TO START WITH PRESENT DAY, AND I WANT TO FRAME THIS IN THE FOLLOWING WAY. SO WEST EURASIANS, WHICH MEANS A REGION THAT EXTENDS FROM EUROPE, ON THE ATLANTIC COAST OF EUROPE, ALL THE WAY TO RNA AND CENTRAL ASIA, IN THE MIDDLE OF ASIA, ARE GENETICALLY VERY SIMILAR, SO IF YOU MEASURE THE SQUARE OF THE FREQUENCY DIFFERENCE, THAT GENETIC VARIANCE, IT'S NOT VERY DIFFERENT. THIS IS ACTUALLY WHAT GAVE RISE IN THE 18TH CORRECT RETO SCIENTIFIC CLASSIFICATIONS OF PEOPLE INTO DIFFERENT GROUPS, THE SORT OF OBSERVATION TO PEOPLE THAT GENETICALLY ARE SIMILAR ACROSS THIS REGION. IF YOU LOOK AT SOME GENETIC VARIANTS, YOU CAN SEE THIS. SO THIS IS A PIE CHART, FREQUENCY OF VARIANTS THAT SOME PEOPLE OF ONE TYPE AND -- THESE ARE VERY UNUSUAL VARIANTS DIFFERENT IN FREQUENCY, SO YOU SEE EURASIA CLUSTERING, AND EAST ASIANS FROM KOREA TO INDONESIA, AGAIN, THERE'S GENETIC SIMILARITY THEN A BIG SHARP DIVISION MOVING INTO WESTERN EURASIA AND IN MUCH OF AFRICA, BUT NOT ALL OF AFRICA, AGAIN, A LOT OF SIMILARITY. AND A QUESTION IS, HOW DID THIS HOPE JE NATE ACROSS THESE REGIONS ARISE? I THINK PEOPLE CENTURIES AGO, WHEN PEOPLE FIRST BEGAN TRYING TO MAKE THESE CAT IMRI, THOUGHT
CATEGORIES
THOUGHT THESE ARE AGE OLD THINGS, SINCE POPULATIONS SEPARATED FROM EACH OTHER OR MAYBE LONGER, IT EVEN ALWAYS EXISTED. IN FACT THE ANSWER IS, THESE ARE ONLY 5,000 YEARS OLD, THESE CLUSTERS THAT WE SEE TODAY, AND FROM ANCIENT DNA, WE KNOW THIS, AND I'M GOING TO TELL YOU HOW WE KNOW THIS IN THE CASE OF EUROPE. I'M GOING TO FIRST TELL YOU ABOUT EUROPE SPECIFICALLY AND I'M GOING TO TELL YOU HOW EUROPEANS TODAY ARE THE RESULT OF MASSIVE MIXTURE OF THREE HIGHLY DIFFERENT POPULATIONS THAT ARE AS DIFFERENT FROM EACH OTHER AS EUROPEAN AND EAST ASIA BUT IT ALL HAPPENED IN THE LAST 9,000 YEARS. SO I'M JUST GOING TO FIRST INTRODUCE A FEW PIECES OF EVIDENCE FROM OUTSIDE GENETICS. THE EVIDENCE OF ARCHAEOLOGY AND THE EVIDENCE OF LANGUAGE. SO WHAT YOU NEED TO KNOW ABOUT THE ARCHAEOLOGY OF EUROPE, THERE'S MANY THINGS, BUT ONE VERY IMPORTANT THING IS THE ARRIVAL OF FARMING IN EUROPE. SO FARMING WAS NOT INVENTED IN EUROPE, IT WAS INVENTED IN THE NEAR EAST BEGINNING BETWEEN 12,000 AND 11,000 YEARS AGO, PROBABLY IN PRESENT-DAY TURKEY OR SYRIA, AND -- OR RAC, AND THE SPREAD FROM THERE, REACHING PRESENT DAY GREECE ABOUT 3,000 YEARS LATER AND IT REACHED BRITAIN AND SCANDINAVIA BY ABOUT 6,000 YEARS AGO, IT SPREAD ACROSS EUROPE DRAMATICALLY, POPULATIONS INCREASED, THE TYPES OF FOOD PEOPLE ATE COMPLETELY CHANGED, THIS WAS A DRAMATIC CHANGE. THE QUESTION HAS ALWAYS BEEN WHETHER THIS MASSIVE TRANSFORMATION ECONOMICALLY WAS ACCOMPANIED BY MOVEMENTS OF PEOPLE, THERE WAS A LOT OF SKEPTICISM THAT THERE MIGHT BE SUCH BUT IT WAS ALWAYS A QUESTION ABOUT WHETHER IT WAS MOVEMENTS OF PEOPLE OR JUST ADAPTION OF IDEAS. WE ALL USE CELL PHONES RIGHT NOW, BUT WE DIDN'T -- WE'RE NOT ALL ETHNICALLY GENETICALLY THE SAME, AND WE ADAPT IDEAS FROM OTHER PEOPLE, SO MAYBE FARMING WAS ADOPTED BY -- FROM PEOPLE'S NAPE NAIBS OPEOPLE'SNEIGHBORS OR
DID IT SP
READ THROUGH PEOPLE. THE OTHER OPTION IS LANGUAGES, SO PEOPLE IN EURM WITH EXCEPTION OF HUNGARIAN, FINNISH -- ENDOEUROPEAN LANGUAGES LEFT A GREAT GAP IN BETWEEN IN THE NEAR EAST AND WE KNOW IT'S BEEN A GAP WHERE IT'S -- WRITING WAS INVENTED IN THE NEAR EAST AND WE KNOW FOR THE LAST -- THERE HASN'T BEEN THESE LANGUAGES IN THIS REGION. SO WHAT IF -- HOW DOES THIS CORRESPOND TO MOVEMENTS OF PEOPLE? SO IN 2009 AND THEN WITH GENOME-WIDE DATA IN 2012, IT BECAME ABSOLUTELY CLEAR FROM GENOME-WIDE DATA THAT THE ARRIVAL OF FARMERS IN EUROPE WAS ACCOMPANIED BY HUGE MOVEMENTS OF PEOPLE. AND THE WAY THIS WAS KNOWN IS THAT PEOPLE SUCCESSFULLY GOT MITOCHONDRIAL DNA, THE DNA IN THE ENERGY FAK TREES OF YOUR SELL, ABOUT 15,000 DNA LETTERS LONG, BUT IT WAS THE PART OF THE DNA THAT FIRST PEOPLE BEGAN TO STUDY WITH ANCIENT DNA BECAUSE THERE'S ABOUT A THOUSAND TIMES MORE COPY PER CELL AND THE CHALLENGE IN DNA IS TO GET ANYTHING AT ALL FROM THESE BONES SO PEOPLE STARTED WITH WHAT THERE WAS THE MOST OF. SO WHEN PEOPLE STUDIED DNA SEQUENCES FROM HUNTER GATHERERS FROM EUROPE, FROM 8,000 YEARS OR 9,000 OR 10,000 YEARS AGO, WHAT THEY FOUND WAS ALL OF THE MITOCHONDRIAL SEQUENCES, A GREAT MAJORITY WERE OF TWO TYPES, WHAT THEY CALLED U4 AND U5, BUT THE FIRST FARMER HS ALMOST NO U4U5, CLEARLY EVIDENT OF PEOPLE COMING IN FROM ELSEWHERE. GENOME-WIDE DATA FROM PEOPLE FROM 5,000-YEAR-OLD -- YEARS AGO IN SWEDEN WAS SUBSTANTIALLY AFTER THE FIRST ARRIVAL OF FARMING, LED BY PONTIFF SCOGLAND AND COLLEAGUES, THEN SHOWED THAT 5,000 YEARS AGO, THE FIRST FARMERS OF NORTHERN EUROPE AND SCANDINAVIA WERE GENETICALLY VERY DIFFERENT FROM THE HUNTER GATHERERS LIVING SIDE BY SIDE, BUT ABOUT AS DIFFERENT FROM EACH OTHER AS EUROPEANS AND EAST ASIANS AND THAT, IN FACT, WERE GENETICALLY VERY SIMILAR NOT TO PRESENT DAY PEOPLE FROM SWEDEN BUT PRESENT DAY PEOPLE FROM SARDINIA AND THEY HAD A MODEL WHERE PEOPLE ARE OF TWO ANCESTRAL POPULATIONS, THE FARMERS WHO HAVE LEFT THE MOST ANCESTRY IN PEOPLE TODAY IN SARDINIA, AND THAT PEOPLE TODAY ARE A MIXTURE OF THESE TWO SOURCES, SO THAT WAS THE MODEL. BUT IN 2012, WE HAD IN OUR LABORATORY ANOTHER OBSERVATION WHICH WAS HARD TO RECONCILE WITH THIS MODEL AND IT WAS THIS: SO WHAT WE FOUND IS WE DEVELOPED A STATISTIC CAL TEST, THE THREE-POPULATION TEST. WE TAKE GENETIC VARIATION DATA, SO WHAT WE'RE LOOKING AT IS HUNDREDS OF THOUSANDS OF POSITIONS IN THE GENOME, WHERE SOME PEOPLE HAVE ONE TYPE AND SOME HAVE ANOTHER TYPE AND WE CAN -- [INAUDIBLE] SO WHAT WE DID IS WE TAKE A SET OF THREE POPULATIONS, A TEST POPULATION THAT WE'RE TESTING TO SEE WHETHER IT'S MIXED AND THEN TWO, POSSIBLE SOURCE POPULATIONS. IF THE TEST POPULATION IS MIXED OF LINEAGES THAT ARE RELATED SOMETIME IN THE PAST TO THE SOURCE POPULATIONS, THE PREDICTION IS, IS THAT THE FREQUENCIES OF THESE GENETIC VARIANTS IN THE TEST POPULATION WILL TEND TO BE INTERMEDIATE BETWEEN THE TWO SOURCE POPULATIONS BECAUSE IT'S MIXED. SO WE CAN TEST THIS OR THE FREQUENCIES TEND TO BE INTERMEDIATE AT ANY ONE POSITION, WE CAN'T MEASURE THAT WITH ANY RELIABILITY, BUT IF WE AVERAGE OVER HUNDREDS OF THOUSANDS OF PHYSICIANS WE CAN OBTAIN A VERY PRECISE DETERMINATION OF WHETHER THE FREQUENCIES ARE ON AVERAGE INTERMEDIATE IN THE TEST POPULATION RELATIVE TO THE SOURCE POPULATION. SO WE HAVE THIS TEST IN HAND, AND WE RAN THIS TEST ON ABOUT 50 WORLDWIDE POPULATIONS, PRESENT DAY ONES, AND WE LOOKED FOR WHETHER EACH POPULATION HAD MIXTURE FROM ANY PAIR OF THE HOAR TWO. SOTHER TWO. SO WHAT WE FOUND IS THAT IN NORTHERN EUROPEANS, PRESENT DAY FRENCH PEOPLE, THERE'S A HUGE SIGNAL OF MIXTURE. ONE OF THEM, THE SOURCE POPULATIONS IS SARDINIAN, CONSISTENT WITH THE IDEA THESE ARE FIRST FARMERS, AND THE OTHER WAS NOT NORTHERN EUROPEANS OR -- BUT IT WAS NATIVE AMERICANS, SO THIS IS A REALLY WEIRD OBSERVATION, AND IT DEFINITELY WAS NATIVE AMERICANS, IT WAS NOT SIBERIAN, IT WAS NOT SOUTH ASIANS, IT WAS NOT EAST ASIANS, THE NATIVE AMERICANS GAVE A MUCH STRONGER SIGNAL. AND WHAT WE PROPOSED WHAT WAS ACTUALLY GOING ON THERE WAS IN NORTHERN EURASIA AT SOME POINT IN THE PAST MORE THAN 15,000 YEARS AGO WHAT WE CALL A GHOST POPULATION, A MAJOR POPULATION THAT LIVED IN NORTHERN EURASIA IN THE PAST, MORE THAN 15,000 YEARS AGO, AND CONTRIBUTED SOME OF THE ANCESTRY OF NATIVE AMERICANS LIVING TODAY AND THAT SOMETIME LATER ALSO CONTRIBUTED ANCESTRY TO EUROPEANS. SO THIS POPULATION DOESN'T EXIST IN A MIXED FORM ANYMORE, IT'S BEEN LARGELY DISPLACED SINCE, EXPLAINING WHY PEOPLE IN SYBERIA TODAY ARE NOT AS GOOD A SOURCE FOR EUROPEAN MIXTURE, AS ARE NATIVE AMERICANS, BUT THE POPULATION EXISTED IN THE PAST. SO WE CALL THEM AIN SHANT NORTH YOU'RE ASIANS AND WE THINK THEY'VE CONTRIBUTED ANCESTRY TO THE AMERICAS AND ALSO NORTHERN EUROPEANS. SO THAT WAS THE PROPOSAL OF A GHOST POPULATION STATISTICALLY RECONSTRUCTED BASED ON PRESENT TADAY POPULATIONS, NO DATA FROM IT. SO WHAT WAS VERY EXCITING TO US WAS AT THE END OF THE NEXT YEAR, THIS GHOST WAS FOUND, ANCIENT DNA GROUP WORKING IN DENMARK, AND THEY SEQUENCED DNA FROM THE 24,000-YEAR-OLD INDIVIDUAL, A BOY FROM THE LATE -- REGION OF EAST CENTRAL SYBERIA. THIS INDIVIDUAL WAS MORE CLOSE LOW RELATED TO EUROPEANS EVEN THAN NATIVE AMERICANS ARE, AND WAS CLEARLY THE CORRECT SOURCE POPULATION FOR THE ANCESTRY BOTH IN THE AMERICAS AND IN EUROPE, TODAY THE DARK BLUE IN THE HEAT MAP SHOWS THERE'S NOT MUCH ANCESTRY LIKE THAT ANYMORE, REFLECTING THE FACT THAT MOST OF THE INDIGENOUS PEOPLE OF SYBERIA TODAY ARE THE RESULTS OF MIGRATIONS THAT HAVE BROUGHT PEOPLE INTO THIS REGION FROM ELSEWHERE MORE CLOSELY RELATED TO EAST ASIANS THAN PEOPLE WERE IN -- AT THAT TIME. SO THIS IS VERY EXCITING TO US THAT WE COULD SEE THE ECHOS OF A GHOST POPULATION RICH IN GENETIC FROM PRESENT DAY PEOPLE AND THAT IT COULD PREDICT ANCIENT DNA, BUT ONCE THAT WAS IN HAND, EVERYTHING BECAME MUCH MORE CLEAR BECAUSE IT'S MUCH EASIER TO LOOK AT IT DIRECTLY THAN TO STUDY IT INCORRECTLY. SO WHAT WE DID, IN WORK LED BY JOSEPH LAZARIDIS WITH JO KRAUSE,
TO COMPARE TO THIS ANCIENT GENOME GENERATED BY THE DANISH GROUP AND TO TRY TO BUILD A POPULATION OF RELATIONSHIP. SO I'M GOING TO NOW DIGRESS FOR 2 MINUTES TO TRY TO TELL YOU WITH HOW WE LOOK AT OUR GENETIC DATA. SO THIS IS DATA FROM 777 PRESENT-DAY INDIVIDUALS FROM WEST EURASIA, SO WEST EURASIA REFERS TO THIS GROUP OF PEOPLE WHO ARE GENETICALLY PRETTY SIMILAR TO EACH OTHER, INCLUDING YOU'RE PEEPS AND CENTRAL ASIANS AND NEAR EASTERNERS, AND PEOPLE -- THE DOTS ARE COLORED BY WHICH POPULATION INDIVIDUALS ARE FROM, BY SELF IDENTIFIED ANCESTRY, ALTHOUGH THE SELF IDENTIFIED ANCESTRY IS NOT USED IN THE MATHEMATICAL TECHNIQUE USED TO DISPLAY THESE POINTS. SO THE NATURE OF THE DATA WE HAVE IS SINGLE NUCLEOTIDE POLYMORPHISM GENOTYPE ARRAY DATA. WHAT THIS MEANS IS THAT WE OR OTHERS HAVE LOOKED AT A -- I'VE TAKEN A GENOTYPING SHIP WHICH INTERROGATES EACH INDIVIDUAL AT ABOUT 600,000 POSITIONS IN THE GENOME. AND SCORES THAT INDIVIDUAL FOR WHAT THEY HAVE, SO AT EACH POSITION THEY WILL HAVE ZERO, ONE OR TWO COPIES OF THE MUTATION THE CHIMPANZEE DOESN'T HAVE, FOR EXAMPLE. AND THERE ARE ABOUT 600,000 ROWS CORRESPONDING TO EACH OF THESE POSITIONS AND 777 COLUMNS CORRESPONDING TO EACH OF THE INDIVIDUALS. WE THEN MULTIPLY THE MATRIX BY ITSELF AND SEE HOW CLOSE EACH INDIVIDUAL IS TO EACH OTHER INDIVIDUAL AND THAT'S A MEASURE OF GENETIC SIMILARITY OF ALL OF THESE INDIVIDUALS AND THEN WE RUN PRINCIPAL COMPONENT ANALYSIS ON THAT AND THAT SEPARATES THE SAMPLES ACCORDING TO THE MOST EFFICIENT WAY OF SEPARATING THEM. SO FOR EXAMPLE, THE FIRST PRINCIPLE COMPONENT MIGHT BE .3 TIMES THE VALUE AT POSITION 1 MINUS .2 TIMES THE VALUE AT POSITION 2, AND PRINCIPLE COMPONENT 2 MIGHT BE SOME OTHER COMBINATION. THE MATHEMATICS WILL EFFICIENTLY SEPARATE SAMPLES IN THIS WAY. SO WHEN WE RUN THIS ON PRET DAY WEST YOU'RE ASIANS, WE SEE A REALLY DRAMATIC PATTERN OF TWO PARALLEL GRADIENT. ALL EUROPEANS ALMOST WITHOUT EXCEPTION FALL IN THIS GRADIENT WITH SAR DINNIANS AT THE BOTTOM END AND NORTHEAST EUROPEANS AT THE OTHER END AND DIFFERENT POINTS INTERMEDIATE, AND ALL NEAR EASTERNERS, ALMOST WITHOUT EXCEPTION, FOLLOW THIS GRADIENT WITH ARMANIANS HERE AND PEOPLE FROM THE SOUTHEAST OVER DOWN HERE. SO THERE ARE VERY FEW INTERMEDIATE POPULATIONS, THE POPULATIONS THAT ARE INTERMEDIATE WITH KNOWN OR POSSIBLY KNOWN RELATIVELY RECENT CONTACTS -- THE ISLAND IN THE MEDITERRANEAN OR JEWISH POPULATIONS. SO WHAT WE DID IS WE TRIED TO LOOK AT THE DATA AND FIND A MODEL OF HISTORY BASED ON GENETIC FREQUENCY SIMILARITY THAT WAS CONSISTENT WITH THE DATA IN OUR RESOLUTION AND I'M NOT GOING TO DESCRIBE THE STATISTICAL TESTS TO DETERMINE THIS MODEL WAS CONSISTENT WITH THE DATA BUT WE WERE ABLE TO SHOW THIS MODEL WAS CONSISTENT WITH THE DATA AS FAR AS WE COULD TELL. SO WE STARTED WITH THESE ANCIENT SAMPLES, THIS 24,000-YEAR-OLD ANCIENT YOU'RE ASIANS AND THIS HUNTER GATHERER, AND MORE DISTANTLY FROM SUB-SAHARAN AFRICANS, SO FAR SO GOOD. NATIVE AMERICANS ARE A MIXTURE OF THESE ANCIENT NORTH YOU'RE ASIANS, AS I SAID, THEY CONTRIBUTED TO NATIVE AMERICANS BUT THEY'RE NOT ALL THE ANCESTRY, THEY ALSO HAVE ANCESTRY FROM EAST ASIANS AND THIS WAS THE MAIN FINDING OF THAT PAPER BY THIS DANISH GROUP, NATIVE AMERICANS TODAY SEEM TO BE CONSISTENT WITH DERIVING FROM A VERY ANCIENT MIXTURE OF TWO GROUPS, THERE'S ALWAYS AN IDEA THAT NATIVE AMERICANS AND EAST ASIANS ARE SOMEHOW SISTER GROUPS THAT DESCREND FOR DISAT THAT PARTICULAR TIMELY FROM WEST YOU'RE ASIANS BUT IN FACT IT'S MORE COMPLICATED AND THIS IS A POTENTIAL RESOLUTION OF THAT. THE FIRST FARMERS OF EUROPE ARE A MIXTURE OF AN ANCIENT GROUP RELATING TO THE HUNTER GATHERERS OF EUROPE AND ANOTHER VERY DISTINCT POPULATION THAT'S ACTUALLY A NEW GHOST POPULATION WE PREDICT FROM THE DATA, WE'VE GOTTEN CLOSER TO FINDING THEM NOW. WE HAVEN'T QUITE FOUND THEM IN TERMS OF ANCIENT DNA, AND IT'S A VERY INTERESTING LINEAGE BECAUSE IT'S SEPARATED FROM THE AN SES INVENTORIES OF WEST AND EAST YOU'RE ASIANS BEFORE THEY SEPARATED FROM EACH OTHER AND PEOPLE IN EUROPE TODAY CAN ONLY BE MODELED AS HAVING MIXTURES FROM ALL THREE OF THESE SOURCES, SO WE HAVE A THREE-WAY, NOT A TWO-WAY MIXTURE. SO THE MIXTURE PROPORTIONS VARY DEPENDING ON WHICH EUROPEAN OP LAITION THEY ARE, BUT ALMOST EVERYBODY HAS SOURCES FROM ALL THREE. SO WHAT ACTUALLY HAPPENED IS THE ANCIENT DNA HAD SHOWN THE FARMER HS COME INTO ANCIENT EUROPE, BUT ALL THE ANCIENT DNA THAT HAD BEEN STUDIED UP TO THIS TIME HAD ONLY SHOWN FARMER OR HUNTER GATHERER ANCESTRY, SO SOMETHING MUST HAVE HAPPENED TO BRING THIS ANCIENT NORTH YOU'RE ASIAN ANCESTRY IN, SO THAT WAS THE NEXT THING WE DID, WE TRIED TO ACTUALLY FIND THEY ENTERED EUROPE. THE WAY WE STARTED THIS IS IN A PAPER BASED ON MITOCHONDRIAL DNA SEQUENCES WHICH LOOKED AT DNA FROM ABOUT 300 MITOCHONDRIAL -- ARCHEOLOGICAL CULTURES TO FIND BASED ON THEIR POTTERY DATING FROM ABOUT 7,500 YEARS AGO TO ABOUT 3,500 YEARS AGO. AND THEY START WITH THE FIRST FARMERS OR EVEN THE HUNTER GATHERERS AND CONTINUE UNTIL THE EARLY BRONZE AGE, WHEN PEOPLE START USING NEW METALS MUCH MORE WIDELY IN EUROPE. SO WHAT WE DEVELOPED IS A TEST OF CONTINUITY, IS IT POSSIBLE THAT THIS POPULATION, THE SECOND POPULATION, IS CONSISTENT WITH DESCENDING DIRECTLY FROM THE FIRST WITHOUT SUBSEQUENT MIXTURE AND SO ON. SO WE HAD A GRID WHERE WE TEST FOR CONTINUITY FOR EACH OF THE SUCSUCCESSIVE POPULATIONS, AND THERE'S A COLORED BOX HERE, WHEN YOU SEE A LINE OF VERY DARK COLORED BOXES SHOWING THE HUNTER GATHERERS ARE CONTINUOUS WITH ALL THE SUBSEQUENT -- THIS IS THE EVIDENCE OF FARMERS REPLACING HUNTER GATHERERS. IT'S VERY DRAMATIC, BUT AFTER THE FIRST FARMERS ARRIVE, ABOUT 7500 YEARS AGO IN THIS PART OF EUROPE, THERE'S CONTINUITY FOR THE NEXT 2 1/2 TO 3,000 YEARS, THERE'S NO VERY STRONG EVIDENCE OF NEW MAJOR INPUT. BUT THEN BANG, 4500 YEARS AGO, THERE'S DISCONTINUITY AGAIN, BEGINNING WITH THE -- CULTURE AND EVER AFTERWARD THERE'S DISCONTINUITY. SO THIS IS A HINT OF WHAT HAPPENED, THE WAY WE DEALT WITH -- THE WAY WE EXPLORE THIS FURTHER IS WE COLLECTED GENOME-WIDE DATA PLUS ADDITIONAL INDIVIDUALS, MANY OF THEM FROM CENTRAL EUROPE, GERMANY, MANY FROM SPAIN, AND THE WHOLE IMPORTANT SERIES OF SAMPLES FROM THE SAMARA REGION OF RUSSIA, WHICH IS FAR EASTERN EUROPE, AND IT DATES TO JUST BEFORE THE -- PERIOD. SO NOW I'M GOING TO ACTUALLY GIVE YOU A SORT OF STILL MOVIE OF WHAT HAPPENS OVER TIME IN EUROPE. THIS IS THE SAME PICTURE I SHOWED YOU BEFORE, PRINCIPAL COMPONENTS ANALYSIS. REMEMBER HERE IS THE NEAR EAST, HERE'S EUROPE, I'M NOW GOING TO GRAY OUT THESE DOTS SO YOU CAN FOCUS ON THE ANCIENT SAMPLE. SO HERE'S THE NEAR EAST, HERE'S EUROPE, AND THE QUESTION; HOW DO THE ANCIENT SAMPLES FALL RELATIVE TO PRESENT DAY POPULATIONS OVER TIME. SO IF YOU LOOK AT -- HUNTER GATHERERS FALL BEHIND EUROPE -- WHAT THIS IS SAYING IS THAT ALL OF THESE HUNTER GATHERERS FROM SWEDEN HERE, FROM WESTERN EUROPE HERE AND SPHAI SPAIN AND
LUXEMBOURG ARE FROM A POPULATION THAT DOESN'T EXIST IN EUROPE IN UNMIXED FORM ANYMORE BUT THAT EUROPEANS TODAY HAVE ANCESTRY FROM THEM BECAUSE THEY ARE MOVED AND SHIFTED IN THAT DIRECTION RELATIVE TO THE NEAR EAST, WHICH YOU CAN ACTUALLY SHOW FORMALLY USING MODELING ANALYSIS. IN THE FAR EAST OF EUROPE, YOU ACTUALLY HAVE POPULATIONS LIKE THIS ON THE EUROPEAN -- ON THE WEST EUR -- ASIAN STEP POINTING AT THE HIGH END OF THIS GRADIENT SIMILAR TO THAT ANCIENT NORTH EURASIAN, AND CARRY THIS ANCESTRY. SO EUROPEANS TODAY ARE ON A GRADE YEPT OF DIFFERENT PROXIMITY TO -- YOU CAN SEE BOTH GRADIENTS REPRESENTED HERE, THE HUNTER GATHERERS RESPONSIBLE FOR THIS SHIFT AND THE AIN SHEPT NORTH YOU'RE ASIANS BY THIS GRADIENT. SO NOW COMES FARMERS, THERE'S A LARGE CLUSTER WE HAVE HERE FROM ALL OVER EUROPE, PRESENT DAY EUROPEANS EXCEPT FOR SARDINIANS POSITION IN A PLATE NOT YET REPRESENTED BY ANY ANCIENT SAMPLES. AT THE SAME TIME, WHAT HAPPENS AFTER THAT IS THE SLIGHTLY LATER FARMER SHIFLTS SLIGHTLY TOWARDS THE HUNTER GATHERERS. THE FARMERS WHO ORIGINALLY ARRIVED WERE SOCIALLY SEPARATED FROM THE HUNTER GATHERERS BUT OVER THE NEXT 2,000 YEARS, THEY BEGAN TO SPEAK TO THEM. WE CAN SEE THAT HAPPENING IN PARALLEL EVERYWHERE. SIMULTANEOUS WITH THIS IN FAR EASTERN EUROPE, ANOTHER MIXTURE OCCURS, PEOPLE FROM THE NORTHERN NEAR EAST, FROM PLACES LIKE AR MEAN YA OR RNA AND FORM A MIXED POPULATION HERE ANALOGOUS TO THESE MIXED POPULATIONS, BUT AT THAT TIME, 5,000 YEARS AGO, WE STILL DON'T HAVE POPULATIONS LIKE BRITISH PEOPLE TODAY, LIKE NORTHEAST EUROPEANS, EVEN LIKE MOST SOUTHERN EUROPEANS. SO WHEN DOES THAT HAPPEN? IT HAPPENS 4500 YEARS AGO WITH THE SUCCEEDING GROUPS. SUDDENLY THESE GROUPS LIKE THIS AND THIS MIXED TOGETHER, AND WE CAN SHOW THESE INFORMAL STATISTICAL TECHNIQUES THAT IT'S VERY DIFFICULT TO GET THIS MIXTURE TO HAPPEN WITHOUT GROUPS VERY CLOSELY RELATED TO THESE AND VERY CLOSELY RELATED TO THESE. ITS PEOPLE ARE WHAT ARE CALLED -- THEY ARE HERDING SHEEP ON THE STEPS. THEY TOOK THEIR SHEEP OUT ON TO THE STEP FOR THE FIRST TIME AFTER DPINNING TO WHEELED VEHICLES, CARTS AND HORSES, WHICH ALLOWED THEM TO MOVE OUT INTO MOBILE HOMES ON TO THE STEP, AND THEY TOOK ADVANTAGE OF THIS LANDSCAPE WHICH WAS UNEXPLOITED BEFORE AND EXPANDED VERY DRAMATICALLY. THERE HAD BEEN A CONTROVERSY ABOUT HOW THEY AFFECTED EUROPE AND WHAT THE GENETIC DATA SHOWS IS THAT THEY ACTUALLY HAD A MAJOR GENETIC IMPACT ON EUROPE AT THIS TIME. SO JUST SUMMARIZING THE POPULATION CHANGES AFTER THE ADD VENT OF AGRICULTURE, IN EUROPE, 8500 YEARS AGO, THE FIRST FARMERS LAND IN THE BALKANS IN GREECE COMING FROM AN ULTIMATE OR GENERAL JEN IN THE NEAR EAST, WE HAVE DNA FROM TURKEY WHICH MATCHES THAT, HYPOTHESES ABOUT WHERE THEY CAME FROM. IF YOU WENT TO THE PROPORTION OF NEAR EASTERN ANCESTRY AND FARMERS AT THIS TIME, IT'S APPROXIMATELY 100% MEASURED BY THIS ORANGE, THE FARMER INDUSTRY. IN THE NEXT 2,000 YEARS. AS THEY MIX, SOCIAL BARRIERS BREAK DOWN, BUT THERE'S STILL NO ANCESTRY RELATED TO THE STAFF AND THAT'S UBIQUITOUS IN EUROPE TODAY. THE NEXT THING THAT HAPPENS 45 HUP YEAR4500YEARS AGO, THEY COME
FROM THE STEPPE AND PEOPLE FROM EUROPE TODAY, ESPECIALLY NORTHERN EUROPEANS, MANY OF THEM HAVE AT LEAST HALF ANCESTRY. IN MANY EUROPEAN POP PLAITION LAITIONS -- YOU MAY ASK, BEFORE I SAID THAT ANCIENT NORTH EUR ASIANS ARE ONLY A QUARTER OR EIGHTH OF THE ANCESTRY OF PRESENT DAY EUROPEANS, BUT WHAT'S GOING ON IS THE ANCIENT ARE -- THE YOU'RE ASIANS WERE TRACER DYE, SO THEY PUT A MINIMUM ON THE PROPORTION OF ANCESTRY COMING FROM THE EAST BUT THEY ACTUALLY BROUGHT A LARGER PROPORTION OF ANCESTRY IN AND THIS IS ACTUALLY PROBABLY THE SINGLE MOST IMPORTANT CONTRIBUTOR AS LONG AS EARLY FARMERS TO EUROPEANS TODAY, THESE PEOPLE. THIS IS NOT ONLY OUR WORK, ANOTHER GROUP PUBLISHES AT THE SAME TIME, THIS DA DANISH GROUP
I MENTIONED BEFORE CAME TO SIMILAR COCONCLUSIONS OF A DRAMATIC TURNOVER IN EUROPE AT THIS TIME. SO NOW I'M GOING TO EXPAND A LITTLE BIT MORE BROADLY IN WEST EURASIA AND TALK ABOUT THE BROADER WEST EURASIAN CONTEXT IN EUROPE. FLOWCIAL SO INTERESTING PLACE BUT IT'S A BROADER PART OF THIS BROAD REGION WHERE TODAY PEOPLE ARE VERY SIMILAR, AS I MENTIONED AT THE BEGINNING, PEOPLE IN WEST EURASIA TODAY ARE GENETICALLY STRIKINGLY SIMILAR. SO WAS IT ALWAYS THAT WAY? SO WE JUST PUBLISHED A PAPER A COUPLE OF MONTHS AGO WHERE WE GOT DNA FROM 44 ANCIENT SAMPLE FROM THE ANCIENT NEAR EAST RANGING FROM 14,000 YEARS AGO TO 4,000 YEARS AGO FROM RNA, FROM ISRAEL AND JORDAN, FROM AR MEAN YA, AND THEY INCLUDE PEOPLE FROM BEFORE THE TRANSITION TO FARMING, JUST AFTER THE TRANSITION TO FARMING, THEY INCLUDE PEOPLE FROM THE EAST OF THE RENAL AND THEY INCLUDE PEOPLE FROM THE WEST OF THE REGION, AND THEY INCLUDE PEOPLE MOVING INTO THE METAL AGES AND TO THE COPPER AGE AND THE BRONZE AGE OVER THIS PERIOD. SO WHAT WE DID, SO I'M GOING TO JUST DIGRESS A LITTLE BIT, AND WHAT MADE THIS ANCIENT DNA WORK IN THE NEAR EAST POSSIBLE AND I THINK WHAT'S GOING TO MAKE WORK IN AFRICA AND SOUTH ASIA POSSIBLE TOO IS THIS INNOVATION OF THE PETRUS BONE, A PIECE OF YOUR TEMPORAL BONE, IT'S THE PART OF YOUR TEMPORAL BONE AROUND THE IBER EAR REGION AND IT HAS BEEN REALIZED IN THE LAST YEAR OR SO THAT THE PETROUS -- ANY PART OF THE SKELETON THAT PEOPLE HAVE LOOKED AT. AND SO ANCIENT DNA SCIENTISTS HAVE MOVED TOWARDS PREFERENTIALLY TRYING TO SAMPLE FROM THE P PETROUS BONE, THAT COMBINED WITH THIS ENRICHMENT TECHNIQUE I MENTIONED HAS INCREASED SUCCESS RATES A LOT. HEAT IS BAD FOR DNA PREB PRESENTATION, BUT USING THE PETROUS -- OTHERS HAVE ALSO HAD SUCCESSES IN REALLY EXCITING CONTEXTS. WE TAKE THESE PETROUS BONES, WE USE A SANDBLASTER TO REMAY -- CONTAINS A LOT OF THE APPARATUS OF HEARING, WE GRIND IT, WE MILL IT, WE OBTAIN POWDER AND SUBJECT THAT TO LATER DNA PROCESSING. SO HERE'S THE PICTURE I GAVE YOU BEFORE BUT WITH MORE SAMPLE, WE NOW HAVE ANCIENT SAMPLES ACROSS THIS WHOLE GRADIENT. WE FIND ANCIENT IRANIANS ARE AT THE TOP OF THE GRADIENT, SO THEY DO HAVE DIFFERENT PROPORTIONS OF ANCESTRY. IF YOU MEASURE THE LEVEL OF DIFFERENTIATION OF THE DIFFERENT GROUPS, IT'S ABSOLUTELY HUGE. SO IF YOU USE ST: FST, IT'S RELATED TO THE SQUARE FREQUENCY DIFFERENCE BETWEEN PAIRS OF POPULATIONS, ALL THESE FOUR POPULATIONS AT THE EXTREMES OF THE QUADRANT, THE FIRST FARMERS TO HUNDRED DER GATHERERS OF WESTERN EUROPE. IRANIANS AND THE FIRST FARMERS OF THE WESTERN PART OF THE NEAR EAST, SO WEST EURASIA 10,000 YEARS AGO HAS FOUR GROUPS THAT ARE AS DISTINCTIVE FROM EACH OTHER AS EUROPEANS AND EAST ASIANS AND TODAY WHA WE CAN TRACK IT
THROUGH TIME. SO THIS TIME IN THE PAST, 10,000 YEARS AGO, THE DIFFERENCE IS AS LARGE AS EUROPEANS AND EAST ASIANS, SO WHAT HAPPENED? 10,000 OR SO YEARS AGO, YOU HAVE THESE FOUR APPROXIMATE SOURCES OF ANCESTRY IN WESTERN EURASIA, EASTERN HUNTER GATHERERS AND GROUPS RELATED TO THEM, FIRST FARMERS FROM THE WESTERN NEAR EAST, FIRST FARMERS FROM THE LATER NEAR EAST, THEY MIXED DRAMATICALLY AND THE REASON DIFFERENTIATION AMONG THE GROUPS TODAY IS SOLO IT'S BECAUSE OF THE MIXTURE, BECAUSE THEY'VE ALL MIXED WITH EACH OTHER. SO THE NEAR EASTERN FARMING EXPANSION AND ITS AFTER EFFECTS HOMOGENIZED -- TO A REGION OF BROAD HOMOGENEITY WHICH LOOKS SO SIMILAR TODAY AND A LOT OF PEOPLE SORT OF IMAGINE IT'S ANCIENT BUT IN FACT IT'S A VERY RECENT PHENOMENON. HYPOTHESIS IS THAT EAST ASIAN AND AFRICAN AGRICULTURAL EX-PAPTIONS MAY HAVE SIMILARLY HHOMOGENIZE THOSE CAT TBRIS WE HAVE IN OUR HEAD, BUT THERE'S NO ANCIENT DNA THAT'S COMPARABLE TO WHAT'S IN EUROPE YET. THE FINAL PART OF MY TALK, I WANTED TO TALK ABOUT THE INSIGHT ABOUT BIOLOGY THAT ARE COMING FROM ANCIENT DNA AND STUDIES OF MODERN DNA THAT RELATED TO -- AND THAT I THINK ARE EXCITING. SO FIRST I INTRODUCE THE IDEA OF A SELECTIVE SWEEP, WHEN A SINGLE MUTATION OCCURS IN A PERSON THAT'S ADVANTAGEOUS THAT MIGHT ALLOW YOU TO DIGEST COW'S MILL NOOK ADULTHOOD, FOR EXAMPLE, IF COW'S MILK IS AVAILABLE AS A SOURCE OF NUTRITION, IT OCCURS IN ONE PERSON AND THEN THE IDEA IS THAT IT RISES TO HIGH FREQUENCY IN A POPULATION UNDER THE PRESSURE OF NATURAL SELECTION. THAT'S WHAT'S CALLED A SELECTIVE SWEEP AND IT'S REALLY EASY TO DETECT USING GENETIC PATTERNS BECAUSE WHAT HAPPENS AROUND THE SELECTIVE SWEEP IS EVERYBODY DESCRENDZ FROM THAT INDIVIDUAL WHO HAS THAT MUTATION PRETTY RECENTLY IN TIME, AND THAT LEAVES A SCAR IN THE GENOME AND THE PATTERN OF VARIATION IS EASY TO DETECT. SO HERE'S AN EXAMPLE IF YOU SCREEN ACROSS THIS CHROMOSOME 2 AND LOOK FOR PLACES IN THE GENOME THAT HAVE STRONG REDUCTION IN VARIATION MEASURED IN ONE WAY OR ANOTHER, YOU'LL SEE A BIG PEAK THAT ALLOWS YOU TO ADJUST COW'S MILK INTO ADULTHOOD. BUT ACTUALLY MOLLY AND COLLEAGUES SHOWED -- HAVE SHOWN THAT, IN FACT, SELECTIVE SWEEPS ACTUALLY ARE RELATIVELY RARE IN HUMAN EVOLUTION, THERE'S A LOT OF EVIDENCE FOR NATURAL SELECTION BUT SELECTIVE SWEEPS HAVE NOT DRIVEN MOST OF IT. INSTEAD MOST OF IT SEEMS LIKELY TO BE SELECTION ON PRE-EXISTING VARIATION IN THE POPULATION THAT SUDDENLY UNDER A NEW ENVIRONMENTAL CONDITION BECOMES ADVANTAGEOUS, AND THAT WON'T LEAVE A SCAR IN THE GENOME. SO IN A WAY WE'RE ONLY FINDING WHAT WE CAN EASILY FIND. THE LOW HANGING FRUIT, WE'RE NOT ACTUALLY FINDING MOST OF WHAT MATTERS. SO WHAT WE ACTUALLY DID WITH OUR ANCIENT DNA DATA IS WE TOOK ADVANTAGE OF THE FACT THAT IT GIVES YOU TIME SERIES DATA. WE UNDERSTAND AS EXPERIMENT LISTS THE POWER OF BEING ABLE DO AN EXPERIMENT AND WATCH THEM DEVELOP OVER TIMES BUT IN HUMAN, UNTIL ABE SHENT DNA, WE HAD TO BE STUCK WITH PEOPLE IN THE PRESENT DAY AND WE HAVEN'T BEEN ABLE TO WATCH HOW PEOPLE EVOLVE INTO EACH OTHER -- SO WHAT WE CAN DO WITH ANCIENT DNA IS WATCH FOR EXAMPLE EUROPEAN POPULATION AND HOW THEY CHANGE OVER TIME, SO WHAT WE DID IS TOOK THIS MODEL FIT INTO THE DATA OF EUROPEANS TODAY BEING A MIXTURE OF THREE HIGHLY DIFFERENT ANCESTRAL POPULATIONS AND THEN WE TOOK EACH POSITION IN THE GENOME WHEN WE ASKED DOES THAT POSITION IN PRESENT DAY EUROPEANS FROM DIFFERENT POPULATIONS HAVE THE FREQUENCY YOU WOULD EXPECT BASED ON THE WHOLE GENOME. SO WE RUN THIS ACROSS THE GENOME GENOME, NOT FOR AFFECTING DIABETES OR NOT BUT WHETHER IT'S CHANGED IN FREQUENCY MORE THAN YOU'D EXPECT BY CHANCE AND WE FIND AT LEAST 12 GENOME-WIDE SIGNIFICANT LOCATIONS WITH THE DATA THAT WE HAD AVAILABLE AT THE TIME THAT WE DID THIS SCAN. SOME OF THEM ARE KNOWN VARIATIONS AND SOME OF THEM ARE ONES THAT HAVE NOT BEEN IDENTIFIED BEFORE. I'LL GIVE YOU A FEW EXAMPLES SO THIS IS LAC LACTASE -- IN ALL SAMPLES RANGING FROM ABOUT 8,000 TO 4,000 YEARS AGO, IT'S VERY RARE, AND TODAY IN NORTHERN EUROPEANS, IT'S VERY COMMON, SO YOU CAN SEE WHY THERE'S VERY STRONG EVIDENCE THAT HAS CHANGED IN FREQUENCY AND REALLY IT ONLY BEGINS TO BE SEEN IN EUROPE ABOUT 4,000 YEARS AGO. SIMILARLY FOR LIGHT SKIN, LIGHT SKIN IS LOWER IN FREQUENCY IN ALL OF THESE ANCIENT POPULATIONS THAT INCLUDE THE SOURCES OF NORTHERN EUROPEANS THAN IT IS IN NORTHERN EUROPEANS TODAY AND EVEN IN SOUTHERN PURE PEENS, THE MAIN VARIANTS AFFECTING LIGHT SKIN SEEM TO HAVE COME IN FROM THE MOST PART FROM THE ANCIENT NEAR EAST WITH THE FARMERS. BLUE EYE COLOR IS REALLY FROM THE INDIGENOUS HUNTER GATHERERS OF EUROPE AND THERE SEEMS TO HAVE BEEN REPEATED TYPES OF SELECTION ON THIS VAIR YAPT. FOR US THIS WAS NOT THE MOST EXCITING TYPE OF WORK, FOR ME THE MOST EXCITING TYPE OF WORK WAS THE GENETIC -- OUR ABILITY TO LOOK AT THIS DATA AT SELECTION FOR COMPLEX TRAITS, FOR THINGS THAT ARE NOT UNDERPIBBED BY A SINGLE GREEN BUT MANY GENES, SO WHERE WE HAD OUR MOST SUCCESS IN THIS AREA AND WHAT I'M MOST EXCITED ABOUT FOR THE FUTURE IS IN THE CASE OF HEIGHT, SO HEIGHT IS A TRAIT THAT'S BEEN STUDIED IN HUNDREDS OF THOUSANDS OF PEOPLE NOW IN MEDICAL GENOME-WIDE ASSOCIATION STUDIES, SO WE'VE ANNOTATED AS A COMMUNITY MILLIONS OF POSITIONS IN THE GENOME FOR HOW THEY AFFECT PEOPLE'S HEIGHT. SO WE HAVE THIS ANNOTATION THAT WE CAN THEN CORRELATE TO CHANGES IN FREQUENCY OVER TIME. SO THERE'S CLEARLY A GENETIC UNDERPINNING TO HEIGHT IN EUROPE, THE ORDER IN WHICH POPULATIONS OF ORDER BY HEIGHT HAS REMAINED THE SAME WITH RELATIVE SMALL STATURE IN PORTUGAL, HIGH STATURE IN CROATIA AND NETHERLANDS, FOR EXAMPLE, AND THAT'S CHANGED OVER TIME -- THAT'S REMAINED CONSTANT OVER TIME. IN 2000, LANGO-ALLEN AND COLLEAGUES PUBLISHED A VERY IMPORTANT PAPER WHERE THEY FOUND 180 INDEPENDENT POSITIONS IN THE GENOME THAT AFFECTS PEOPLE'S STATURE. A LATER PAPER BY JOEL -- FOWNDZ THERE'S SELECTION FOR DIFFERENT STATURE, THEY FOUND CONSISTENT EVIDENCE THAT THE VARIANTS AFFECTING TALLER STATURE, LARGER STATURE, WERE MORE COMMON IN NORTHERN THAN IN SOUTHERN EUROPEANS AND THAT THIS WAS TOO MUCH TO BE EXPLAINED BY RANDOM FLUCTUATION SO MUST HAVE BEEN CALL FOR SMALLER OR TALLER BUT THEY DIDN'T KNOW WHICH. SO WE HAD ANCIENT DNA FROM EACH OF THESE ANCIENT POPULATIONS AND WE COULD SCORE THEM FOR WHETHER THEY HAD -- WHAT THEIR PREDICTED HEIGHT WAS BASED ON THE GENETIC VARIANT THAT FROM PRESENT DAY GENOME WIDE ASSOCIATION STUDIES, NO AFFECT IN HEIGHT. SO AS POPULATIONS MOVE INTO THE MIDDLE OF THE FIRST FARMING PERIOD, THERE EMERGES A DIRNS BETWEEN NORTHERN EUROPEAN FARMERS AND SOUTHERN EUROPEAN FARMERS AND THERE'S A STATISTICALLY SIGNIFICANT DROP IN GENETICALLY PREDICTED STATURE IN SOUTHERN EUROPEANS. SO WHAT THIS MEANS IS THAT THE REDUCED STATURE IN SOUTHERN EUROPE IS ACTUALLY DO AND DRIVEN IN LARGE PART BY SELECTIONS TO REDUCE STATURE, THERE'S A SEPARATE EFFECT IN THE STEP WHERE THERE SEEMS TO HAVE BEEN SELECTION TO PROBABLY INCREASE STATURE IN THE STEPPE, THESE PEOPLE COME IN TO EUROPE 4500 YEARS AGO. THAT'S WHAT'S GOING ON, THAT'S WHAT'S EXPLAINING IT. IT'S NOT CLEAR WHY TALLER OR SHORTER STATURE WOULD HAVE BEEN FAVORED IN CERTAIN ENVIRONMENTS AND SO I THINK THAT THIS IS REALLY AN EXCITING WAY TO INTERROGATE HOW TRAITS EVOLVE OVER TIME, AND I THINK IT COMBINES THE POWER OF GENOME-WIDE ASSOCIATION STUDIES WHICH HAVE NOW BEEN CARRIED OUT FOR MANY TRAITS, DIABETES AND HEART DISEASE AND MANY CANCERS AN AUTOIMMUNE DISEASES AS WELL AS HEIGHT AND OTHER TRAITS, WHICH HAVE RURLTE RESULTED IN ANNOTATIONS OF THE GENOME, OFTEN THEY'RE VERY, VERY MINOR EFFECTS BUT WE CAN CORRELATE THEM TO CHANGES IN FREQUENCY OVER TIME TO SEE WHETHER THERE'S BEEN SELECTION IN ONE WAY OR ANOTHER WITH CERTAIN TRAITS. HEIGHT IS THE BEST UNDERSTOOD TRAIT BUT THE PAPER BY JONATHAN PRITCHARD'S GROUP THAT'S ON A PRE-PRINT SERVER AND I HOPE WILL BE PUBLISHED SOON FINDS EVOLUTION NOT JUST OF HEIGHT, WHICH IS A STRONG SIGNAL, BUT OF INFANT HEAD CIRCUMFERENCE AND INSULIN LEVELS, AGE OF PUBERTY, ESPECIALLY IN FEMALES, YOU CAN SEE EACH OF THE TRAITS IN BRITAIN WHERE THE STUDY IS BASED IN THE LAST 2,000 YEARS IN THE SAME TYPE OF WAY, BY CORRELATING CHANGES IN FREQUENCY INFERRED FROM GENETIC DATA WITH THE ANNOTATIONS FROM GENOME-WIDE ASSOCIATION STUDIES. SO I WANT TO FINALLY CONCLUDE BY ASKING THE QUESTION, IF YOU GO BACK IN TIME, HOW MODERN TRAITS APPEAR SO I'M GOING TO GIVE YOU A LITTLE BACKGROUND ON THIS. FIRST ANATOMICALLY MODERN HUMANS, PEOPLE WITH SKELETONS SIMILAR TO HOURS DATED ALMOST 200,000 YEARS AGO IN EAST AFRICA, AND THE ARCHEOLOGICAL STUDIES OF THESE PEOPLE, THE STONE TOOLS THEY MADE, THE ORNAMENTS THEY LEFT BEHIND, INDICATE THERE HAVE NOT -- THAN THEIR CAN YOU SIPS IN EUROPE, OR ELSEWHERE IN ASIA OR ELSEWHERE IN AFRICA. THAT IS, THESE PEOPLE WERE ANATOMICALLY LIKE US BUT THEY WERE BEHAVIORALLY JUST LIKE NEANDERTHALS AND MADE SIMILAR TOOLS, AND IT TOOK 150,000 YEARS FOR THESE PEOPLE TO ACTUALLY BECOME BEHAVIORALLY MUCH MORE LIKE US. ARC LOGICALLY THERE'S DOCUMENTED ABOUT 50,000 YEARS AGO A QUICKENING OF ARCHEOLOGICAL CHANGE WHERE MUCH MORE COMPLEX TOOLS START TO BE MADE INCLUDING BONE TOOLS, NOT JUST STONE TOOLS, LEFT BEHIND, EVIDENCE OF CLEARING, ORNAMENTATION, DELIBERATE BURIALS, ALL OF THESE OCCURRED AT SOMETIME BEFORE BY PERHAPS NOT AS INTENSE. WHAT YOU'RE ACTUALLY SEEING HERE IS EVIDENCE OF GREAT CHANGE IN THE HUMAN POPULATION AND PERHAPS IT'S EVEN A BIOLOGICAL CHANGE, MAYBE THERE'S BEEN A GENETIC SWITCH, A CHANGE MAYBE IN NEUROLOGICAL GENE THAT WOULDN'T HAVE LEFT AN IMPRINT IN THE BRAIN CASE BUT ROSE TO HIGH FREQUENCY AND ALLOWED COMPLEX BEHAVIORAL -- MAYBE LANGUAGE AND THAT POPULATION EXPANDED LARGELY DISPLACING ALL THE PEOPLE WHO WERE THERE, THEY PROBABLY EXPENDED WITHIN AFRICA TOO, DISPLACING MOST OF THE POPULATIONS THAT WERE THERE, THAT WAS THE IDEA THAT WAS VERY ECEXCITING. THIS IDEA OF A GENETIC CHANGE THAT MIGHT EXPLAIN FROM OUR RELATIVES HAS BEEN INVOKED BY A LOT OF PEOPLE, INCLUDING ME, TO JUSTIFY SIR CHS FOR EVOLUTIONARY STUDIES, OR WE'RE GOING TO FIND THE KEY TO WHAT MAKES US SPECIAL, BY SEQUENCING THE DIFFERENCES BETWEEN US AND CHIMPANZEES, WE'RE GOING TO READ IT LIKE A BOOK AND FIGURE OUT WHAT MAKES US SPECIAL. THE PROBLEM IS, WE STILL HAVEN'T FIGURED OUT ANYTHING ABOUT WHAT'S SPECIAL, EVEN THOUGH THEY'VE RESULTED IN SUCH AMAZING SUCCESSES IN OTHER WAYS. SO WHAT I WANT TO CONCLUDE ON IS SOME WORK THAT SPEAKS TO SOME OF THESE QUESTIONS, THIS PAPER THAT FRANCIS REFERRED TO JUST BEFORE, WHICH IS A STUDY WE JUST PUBLISHED OF 300 HIGH QUALITY GENOME SEQUENCES FROM 142 DIVERSE POPULATIONS FROM PLACES SHOWN IN THIS MAP AND CONTAINS A LOT OF RESOURCES ESPECIALLY FROM POPULATIONS THAT HAVE NOT BEEN SAMPLED BEFORE, FOR EXAMPLE, SOME POPULATIONS FROM AFRICA OR NEW GUINEA AND ELSEWHERE. THE FIRST THING THAT WE CAN DO IS WE CAN LINE UP THE GENOME SEQUENCES FROM THE MOTHER AND FATHER, COUNT THE NUMBER OF DIFFERENCES BETWEEN THEM, AND THEY PROVIDE A MEASURE OF TIME. SO WE CAN ASK HOW LONG HAS IT BEEN SINCE THE WERE -- AND YOU CAN COMPARE TWO POPULATIONS TO EACH OTHER AS WELL, AND SO WHAT WE'VE DONE IS WE'VE USED THE TECHNIQUE DEVELOPED BY ANOTHER GROUP FOR ESTIMATING POPULATION SEPARATION, AND FOR MANY PAIRS OF POPULATION, MUCH OF THE SEPARATION HAPPENED BEFORE 50,000 YEARS AGO, WHICH IS WHEN THIS GREAT CHANGE OCCURS. SO THE FIRST THING TO KEEP IN MIND IS THAT THESE BIG CHANGES THAT DOCTORS OCCURRED HAPPENED BEFORE THE SEPARATION OF A LOT OF PRESENT DAY POPULATION SO IF YOU BELIEVE THAT THERE WAS A BIG CHANGE, IT MUST HAVE OCCURRED MORE IN THE ANCESTRY BUT ALMOST ALL POPULATIONS TODAY ARE CAPABLE OF FULLY -- >FULLY --THE SECOND RESULT, WE USED THE SAME TECHNIQUE TO ESTIMATE AT EACH FIRST PLACE IN THE GENOME HOW FAR BACK IT IS, BUT JUST COUNTING DENSITY MUSHARRAF TAITIONS. SO THIS IS A RESULT ON A CHROMOSOME FOR PERSON ONE, IS HOW FAR BACK IN TIME DO THE TWO CHROMOSOMES SHARE A COMMON ANCESTRY? HOW FAR BACK IN TIME DO WITH VERY TO GO FOR EVERYBODY -- THE ANSWER IS THERE'S REALLY NOWHERE IN THE GENOME WHERE SOMEBODY SHARES SL -- MOST OF IT WE'VE COVERED. AND IF YOU BELIEVE THIS GENETIC SWITCH IDEA, YOU MUST BELIEVE THERE'S A GENETIC -- ROSE HIGH FREQUENCY BY ABOUT 50,000 YEARS AGO, AND THEN EVERYBODY TODAY. BUT THAT'S NOT WHAT WE SEE AT ALL, THERE'S NOTHING EVEN PLAUSIBLY CLOSE TO THAT, SO THE CONCLUSION OF THIS PART OF THE TALK IS THAT I THINK THAT FOR ME, IT'S SUGGESTING THAT THERE'S NOT LIKELY TO BE SIMPLE EXPLANATIONS FOR BEHAVIOR, EVOLUTION OF THESE BEHAVIORALLY COMPLEX TRAITS THAT ARE EVIDENT IN THE ARCHEOLOGICAL RECORD AND MAYBE YOU MIGHT GENERALIZE MORE BROADLY IF THERE MIGHT NOT BE SIMPLE EVOLUTIONS OF HUMAN TRAITS. WE DON'T THINK THERE HAVE BEEN ANY EVIDENCE IN ENABLING MUTATION BUS SUDDENLY APPEARED AND ALLOWED THEM TO THINK THAT IT WAS SOMEHOW A POPULATION, WHAT WE CALL A MUTATION LIMITED STATE, WHICH COULD DO THINGS BEFORE AND THEN SUDDENLY AN AMAZING MUTATION HAPPENED THAT ALLOWED PEOPLE TO DO THINGS BECAUSE WE DON'T SEE ANY EVIDENCE IN THE MUTATION THAT IT COULD HAVE BEEN LIKE THAT. I THINK THAT DISTINCTTIVELY HUMAN BEHAVIOR BECAUSE OF THIS, IT SEEMS MORE LIKELY TO BE A COMPLEX TRAIT, NOT LIKE SKIN COLOR OR LACTASE PERSISTENCE, THAT HUMAN POPULATIONS BEFORE WITH THEM NECESSARY TO DO EVERYTHING WE DO AND IT SEEMS IN THIS CONTEXT THAT MAYBE ENVIRONMENTAL AND CULTURAL CHANGES WHICH PERHAPS -- LIKELY TO HAVE BEEN DRIVERS OF GREAT CHANGES THAT OCCURRED, AND IT'S AN INTERESTING CASE HERE WHERE GENETICISTS SEARCH WHERE GENETICS IS THE EXPLANATION BUT HERE IN A SORT OF PARADOXICAL WAY, GENETIC -- WHAT MAKES US SPECIAL OR DIFFERENT, BUT I ACTUALLY THINK THAT'S AN OPPORTUNITY IN A WAY BECAUSE I ACTUALLY THINK THAT THE INTERESTING INSIGHTS WILL COME FROM TRYING TO EMBRACE THE COMPLEXITY RATHER THAN TO COME UP WITH A SIMPLE EXPLANATION, SO THAT'S THE END OF MY TALK AND I THINK THERE'S TIME FOR A FEW QUESTIONS. [APPLAUSE] >> THANKS, DAVID, THAT WAS FASCINATING INDEED AND I SUSPECT PEOPLE DO HAVE QUESTIONS. PLEASE USE THE MICROPHONE SO THAT PEOPLE LISTENING ON THE VIDEO CAN ALSO HEAR THE QUESTION. PLEASE START RIGHT HERE. >> I'M WONDERING ABOUT DISEASE CORRELATIONS AND WITH YOUR PATTERNS OF THE THREE OR FOUR EUROPEAN BACKGROUNDS, IS THERE ANY LINKAGE TO ANY PARTICULAR DISEASES, IN OTHER WORDS, CAN YOU SAY THAT SOME DISEASES, GENETIC DISEASES ORIGINATED IN THOSE DIFFERENT POPULATIONS? >> THE QUESTION IS WHETHER CERTAIN DISEASES ARE LIKELY TO HAVE ORIGINATED IN ONE POPULATION OR THE OTHER. WE IN FACT TRY TO LOOK AT THE GENOME-WIDE ASSOCIATION CATALOGS AND CORRELATE THEM TO DIFFERENT CONTRIBUTIONS IN ONE WAY OR THE OTHER AND WE HAVEN'T REALLY OBTAINED ANY ADEQUATE RESULTS IN THAT WAY EXCEPT FOR HEIGHT YET. I THINK IT'S IN PART DUE TO THE LIMITATION OF THE ANCIENT DNA DATA WE HAD AVAILABLE AT THE TIME. THIS WAS DRIVEN BY SAMPLE SIZE, SO I THINK IN ORDER TO OBTAIN BETTER INSIGHT, WHAT WE WANT TO DO IS INCREASE THE SAMPLE SIZE AND THAT'S HAPPENING VERY RAPIDLY NOWP, SO I THIN NOW, SO
I THINK COMBINING THE ANCIENT DNA DATA WITH THE -- WE MIGHT BE ABLE TO PARTITION DISEASES IN THIS WAY. >> AT THE BEGINNING OF THE TALK, I GOT THE IMPRESSION THAT THE ENRICHMENT METHODS WERE YOU USING, WERE YOU SAMPLING ACTUALLY A VERY SMALL FRACTION OF THE GENOME BUT GETTING LOTS OF COVERAGE OF IT. WHEN YOU USE THE PHRASE GENOME-WIDE, DOES THAT MEAN JUST THAT ENRICHMENT OR DO YOU LITERALLY MEAN ATTEMPTING TO GET THE ENTIRE GENOME, AND THEN COMING LATE IN THE TALK WHEN YOU TALKED ABOUT YOUR 300 GENOME, AGAIN, WHAT FRACTION OF THE GENOME IS BEING SAMPLED? >> RIGHT. SO THE GENOME-WIDE DATA, WHAT WE'RE USING IN OUR LABORATORY IS AN ENRICHMENT TECHNIQUE THAT DOESN'T SEQUENCE THE WHOLE GENOME FOR MOST SAMPLES. WE SOMETIMES SEQUENCE THE WHOLE GENOME BUT WE ENRICH BASICALLY ABOUT 1.2 MILLION POSITIONS THAT ARE OF INTEREST. FOR -- IT'S ONLY A% OF POSITIONS
5% OF --
BUT THE MEDIAN MIGHT BE .3 AND MANY SAMPLES ARE HEAVILY COVERED AT THOSE POSITIONS BUT NOT EVERY POSITION IN THE GENOME AND THAT'S THE SECRET WE'VE USED FOR VERY LARGE NUMBER OF SAMPLES. THERE IS A POSSIBLE THING ONE COULD DO, WHICH WOULD BE TO SHOTGUN DEEP SEQUENCE A LOT OF THESE SAMPLES AMENABLE TO IT BECAUSE THEY HAVE HIGH PERCENTAGE OF HUMAN DNA. THAT IS NOT WHAT WE HAVE DONE ON THESE SAMPLES SO FAR, IT'S -- THE LAST QUESTION YOU ASKED WAS ABOUT THE 300 GENOMES, THAT'S PRET DAY GENOMES TO -- >> JUST ONE LAST QUESTION WHEN YOU SHOWED THE MOORE'S LAW OF GROWTH OF APRIL SHENT DNA SEQUENCES IN GENERAL FROM MANY INVESTIGATORS, IN GENERAL, ARE MOST OF THOSE SEQUENCES LIKE YOUR, JUST SAMPLING SOME SMALL PERCENT OF THE GENOME, OR ARE ANY OF THOSE A LARGER -- >> OUR LABORATORY HAS PRODUCED MORE THAN HALF OF THE SEQUENCES PUBLISHED TODAY, SO MOST OF THEM PUBLISHED ARE FROM THAT TECHNIQUE. BUT A SUBSTANTIAL FRACTION ARE SHOTGUN SEQUENCE DATA. >> THE GRANTS FROM PRINCETON HAD DONE STUDIES OVER SEVERAL DECADES LOOKING AT DARWIN'S, FINCH, LOOKING AT -- NOTICE THERE'S AN EVOLUTION OF PEAK SIZE BASED ON ENVIRONMENTAL IMPACT. I'M JUST CURIOUS TO KNOW IF YOU LOOK AT THE PARTICULAR POINT IN TIME WHERE YOU SEE THE DIFFERENT GROUPINGS, DO THEY MERGE AND IS THERE SORT OF A BLENDING AND AT TIMES OF GEOLOGICAL STRESS? AND THEN ANOTHER QUESTION I HAD HAD TO DEAL WITH WHETHER OR NOT -- LIKE YOU MENTIONED THE LACTOSE AND ABILITY TO DIGEST MILK. ARE THERE GENE DRIVERS ASSOCIATED WITH THAT SO THAT IT BECOMES MORE AND MORE PREVALENT WITHIN A POPULATION? >> SO -- YEAH, SO I THINK AGAIN WE'RE AT THE POINT WHERE WE DON'T HAVE ENOUGH ANCIENT DNA YET. MOST OF OUR ANCIENT DNA IS FROM A RELATIVELY NARROW TIME SLICE THAT I REPORTED ON, BETWEEN ABOUT 7,000 YEARS AGO AND ABOUT 4,000 YEARS AGO. AND SO IN ORDER TO REALLY CLOCK EUMENDOCUMENT, I WOULD WANT TO
FILL IN A MORE RECENT PERIOD AND ALLOW ONE TO TRACK EXACTLY WHEN THE CHANGES OCCUR. IN SOME APPROXIMATION, WE HAVE TWO TIME POINTS, THE FIRST FARMERS AND THEIR IMMEDIATE SUCCESSORS IN EUROPE AND THE PRESENT. SO THAT'S 2 POINTS AND WE CAN'T REALLY DO MORE THAN DRAW VOLUME CONNECTING THOSE TWO-POINT, BUT WITH MORE DATA, WE CAN ACTUALLY FILL IN THAT LINE AND FIGURE OUT -- WITH LACK TASTE PERSISTENCE, YOU CAN ACTUALLY SEE IN OUR DATA, YOU CAN SEE THE FIRST OCCURRENCES AND SUBSTANTIAL FREQUENCY ABOUT 4 OR 5,000 YEARS AGO. YOUR OTHER QUESTION? >> HAD TO DO WITH GENE DRIVERS. I'M JUST WONDERING, YOU HAVE A MUTATION AND THEN IT SEEMS TO BECOME MORE PREVALENT IN A POPULATION SUCH AS LACTOSE INTOLERANCE. AND I'M WONDERING, IS THERE SOMETHING THAT'S DRIVING THAT PREVALENCE? >> MY GUESS, I DON'T KNOW, BUT I THINK WHAT WE KNOW -- I THINK WHAT WE GENERALLY ASSUME IS THAT THE INCREASE IN FREQUENCY OF LACTOSE PERSISTENCE IS DRIVEN BY THE -- OF MILK, SO PEOPLE HAVE MORE OFFSPRING IF THEY ARE ABLE TO DO THAT, SO IT'S JUST NATURAL SELECTION. I DON'T THINK WE HAVE ANY EVIDENCE, BUT I'M NOT SURE I'M ANSWERING THE QUESTION. >> YOU MENTIONED YOU'RE LOOKING AT A SPECIFIC JEAN, AB YOU'RE NOT LOOKING AT WHAT MAY BE ASSOCIATED WITH THAT GENE, IS THAT CORRECT? >> THAT'S RIGHT. THERE'S A LOT TO DO. >> YOU MENTIONED THE NARROW TIME SLICE OF SAMPLES YOU'RE LOOKING AT. IS THERE ANY POSSIBILITY YOU CAN COMPARE YOUR SAMPLE AND IN PARTICULAR THE GHOST POPULATION TO EXISTING OLDER ANCIENT -- BEYOND THE TALL GENOME AND THEN IT'S OPEN TO SEE IF THOSE IN ANY POSSIBLE WAY ARE RELATED TO -- >> YEAH, I'VE DONE A LOT OF WORK, MOST OF MY WORK IN ANCIENT DNA AND THE RELATIONSHIPS BETWEEN THESE GROUPS. UP WITH THING THAT WE CAN SEE, AND IT'S ACTUALLY IN THIS WORK ON THE NEAR EAST, WE CAN ACTUALLY SEE THERE'S BEEN -- ACTUALLY ANOTHER PAPER WE PUBLISHED A FEW MONTHS AGO WHICH WAS ON UPPER -- EUROPE SO EUROPEANS, 51 SAMPLES BETWEEN 45,000 AND 7,000 YEARS AGO IS THAT THERE'S ACTUALLY BEEN A DECREASE IN NEANDERTHAL ANCESTRY OVER TIME IN EUROPE SO THAT IN THE FIRST EUROPEAN MODERN EUROPEANS WE HAVE DATA FROM OR YOU'RE ASIAN, IT'S ABOUT 5% AND DECREASES TO ABOUT 2% TODAY, SO THERE WAS MORE NEANDERTHAL ANCESTRY IN THE PAST, AND IT'S ACTUALLY INTERESTING BECAUSE THESE POPULATIONS ARE DIRECTLY ANCESTRAL TO SOME OF THE LATER POPULATIONS SO WE THINK THERE'S BEEN NATURAL SELECTION TO REMOVE NEANDERTHAL ANCESTRY, IT'S SLIGHTLY DISADVANTAGEOUS BUT WE CAN ACTUALLY TELL VERY ACCURATELY WHETHER THE PROPORTION OF AN SES TREES IN INDIVIDUAL GENOME -- NONE OF THESE GROUPS HAVE EXTREMELY LARGE AMOUNTS EXCEPT FOR ONE SAMUEL FOUND, WHICH HAD ABOUT 10%, AND WE FOUND THAT -- WAS A NEANDERTHAL SO WE'RE VERY LUCKY WE FOUND A 40,000-YEAR-OLD INDIVIDUAL LIKE THIS. BUT MOST OF THESE SAMPLES ARE MUCH LESS AND NOT PARTICULARLY UNUSUAL FOR THEIR TIME. >> DAVID, LET ME ASK YOU ABOUT WHAT YOU PERCEIVE AS THE MATURITY OF THIS VERY RAPIDLY MOVING FIELD. AMAZING OBSERVATIONS THAT HAVE COME FORWARD IN JUST THE LAST FEW YEARS, BUT WHAT DO YOU THINK IS GOING TO HAPPEN IN THE NEXT HALF DECADE? ARE THERE MANY MYSTERIES YET TO BE REVEALED, OR IS THE BIG PICTURE NOW PRETTY MUCH TAKING SHAPE? >> THANK YOU. I THINK THAT WE'RE REALLY JUST AT THE BEGINNING. THERE ARE SO MANY THINGS TO DO SO I CAN JUST RATTLE OFF A SERIES OF THINGS ONE IN PRINCIPLE CAN DO. WE COULD INCREASE SAMPLE SIZE TREMENDOUSLY AND NOW THAT'S VERY EASY AND WE CAN ACTUALLY REALLY STUDY NATURAL SELECTION WITH REALLY HIGH STATISTICAL POWER. THE STUDY I TOLD BUT WAS JUST A PROOF OF PRINCIPLE, BUT THERE'S REALLY THE POSSIBILITY OF REALLY DOING THIS WITH EXTRAORDINARY ACCURACY OF DOING SCANS OF WHAT NATURAL SELECTION LANDSCAPE WAS LIKE 10,000 YEARS AGO AND COMPARING IT TO THE PRESENT, SELECTION CHANGE AND ITS NATURE. ANOTHER POSSIBILITY IS TO STUDY WHAT HAPPENED AFTER THE BRONZE AGE IN EUROPE FOR EXAMPLE, WHAT TRANSFORMATION HAS HAPPENED SINCE THESE PEOPLE AROSE. EUROPE IS A TINY PART OF THE WORLD. ONE COULD STUDY THE AMERICAS, ONE CAN STUDY EAST ASIA, ONE CAN STUDY SOUTH ASIA, ONE CAN STUDY AFRICA, MANY PARTS OF AFRICA, AND THERE'S ALL OF THESE PLACES TO STUDY AND I DID NOT MENTION ANY PLACES. THERE'S ALSO THINGS THAT ONE CAN DEADO WITH ANCIENT DNA WHICH
WILL ANSWER QUESTIONS THAT HAVE NOT BEEN ANSWERED. WE CAN PROBABLY USE DNA TO ESTIMATE THE CENSUS SIZE OF POPULATION. YOU CAN SEE GENETICALLY IN PEOPLE TODAY THAT -- CHINESE HAVE A VERY LARGE DENS SIZE BECAUSE THEY HAVE RELATIVE FEW CHILDREN SO IF WE CAN GET SUBSTANTIAL -- WE CAN ESTIMATE POPULATION SIZE, SOMETHING WE'VE NEVER DONE, AND THAT'S SO IMPORTANT FOR ECONOMICS, FOR HEALTH EVEN. ANOTHER THAT'S VERY INTERESTING IS THAT THIS BIG ARRIVAL OF PEOPLE FROM THE STEPPE, HERDING SHEEP, COME INTO EUROPE WHERE THEY WERE DENSELY POPULATED BY FARMER, HOW DO THEY MAKE AN IMPACT WHERE PEOPLE THAT ALREADY HAVE AN EFFICIENT WAY OF LIVING CAN HAVE AN IMPACT. THERE WAS A STUDY THAT STUD STUD HE'D THE GENOME SEQUENCES OF THESE SAMPLES AT THE TIME OF THE IMPANGT, AND THEY FOUND OUT FROM THESE FOUR TO 5,000 -- THERE IS PLAGUE BACILLUS, WHICH SOME HAD THOUGHT ONLY JUMPED INTO THE SEVENTH CENTURY, BUT HERE IT IS 5,000 YEARS, IT DOESN'T HAVE THE VIRULENCE FACTORS NEEDED FOR BUBONIC PLAGUE BUT IT DOES FOR PNEUMONIC PLAGUE. 5 TO 10% OF THE SAMPLES THAT THEY STUDIED HAD PLAGUE IN THEIR TEETH AT THE TIME THEY DIED, WHICH MEANS THEY PROBABLY DIED ON IT, AND ONE POSSIBILITY THAT THESE PEOPLE SUGGESTED WHICH IS AN EXCITING POSSIBILITY BUT WHICH SHOWS THE POTENTIAL POWER, EUROPE 5,000 YEARS AGO WERE LIKE MAYTIVE AMERICANS, GENETICALLY NAIVE POPULATION SUDDENLY EXPOSED TO DISEASE WITH THE STEP E AND THERE'S GREAT DISEASES THAT SWEPT ACROSS EUROPE AT THIS TIME AND MADE IT POSSIBLE FOR PEOPLE TO WHO HAVE IN WHO OTHERWISE WOULD NOT HAVE HAD THE DEMOCRATIC NICHE. >> I THINK HE'S RIGHT, THERE'S A LOT OF WORK TO DO HERE. IF YOU WANT TO CONTINUE THIS CONVERSATION WITH DAVID, PLEASE JOIN FOURS COFFE FOR COFFEE AND
COOKIES AN
D MEANWHILE LET'S THANK OUR SPEAKER AGAIN.
