genetics Exam 3

Molecular Markers general

-short DNA segments identified using molecular techniques -- not relying on allelic difference (don't care what specific allele is there) - can use regions of DNA that do not encode genes to be used as markers along chromosome Molecular marker: segment of DNA that is found at specific site along chromosomes and has properties that enable it to be uniquely recognized using molecular tools- like PCR or gel electrophoresis *so can be single nucleotide or segment, like alleles they can be polymorphic why useful? help us determine approx location of unknown gene that causes human disease -- can follow markers of people with diseases and if find common marker in all who have the disease (then know somewhere in or around that marker may be specific disease causing allele!)

why neandrathal and denisovan sequences persisted and what for?

Neandrathal sequences correct hypothesis!!!: deandrathal and Denison sequences persisted in humans likely due to selection!!- mostly in the immune system !!— idea being that If neandtrhals established in all diff regions world they ahave been exposed to pathogens then when humans came in w weak immune systems its beneficial for them to acquire those sequences from them One ex of denisoven gene that persisted: O2 carrying in blood-tibeten populations that are excellent at living in low o2 have all closeslt to denisiovan allele that allows them to persist at high altitude

better way: how targeting mosquitos

OR can target mosquitos make modified versions of mosquitos either 1- that cant take up plasmodiums vs 2 — just eliminate the mosquito pop - not bite us so don't spread malaria or zika etc BUT - hard thing- If you design a mosquito that has a genetic background that makes it die- it will not be fit to pass on genes- hard to do this bc other ones that exist will live and yours will just die study: done w flies! to rid of pop : target ability to reduce Drosophila summary shown here (different exon numbers than in mosquitos) Targeting the female-specific exon in doublesex (exon 5 in mosquitos) will impact female fertility only Well studied series of events in mosquitos that leads to development of diff sexes Gene at top- sex lethal- can be alternatively sliced in either male or female EXPLANATION: there is a specific exon (part of DNA encoding protein) - called double sex sex lethal- this is a protein that influences how this exon is created (how it is spliced in male v female) in females: there is also a presence of a transformer protein - a specific codon 4- is included in the exon so female double sex protein exon: 1234 male double sex is spliced differently bc no transformer present male double sex exon 12356 *NO PRESENCE of 4 the corresponding female doubles protein represses male genes and activates female and vise versa in females- original sex lethal sequences allow for sex lethal protein to be made in male- at 4rd codon is stop so they don't actually make the sex lethal protein w sex lethal sequence **then bc females have the sex lethal protein (its job is to influence splicing so it splices to sequence that encodes transformer protein and transformer protein ENCODES FEMALE SPECIFIC EXON (4) WHICH ALLOWS FOR THEIR VERSION OF DOUBLE SEX males: early stop codon at 2- so no sex lethal no influence on splicing so incorrect splicing= no transformer protein no tranformer protein= diff double sex (bc no 4 codon) what did w CRISPR: Created a CRISPR construct that has a guide mrna— going to add something to exon 5 which is the female specific exon in double sec In gray is guide RNA from CRISPR- will guide cas9 to cut at intron/exon border Create double stranded break at that location then when DNA repair solves that break- will leaf to insertion or deltion- which will mess up splicing- repair by NHEJ and — just pasted back together and in mean time some lost or addition- but just trying to screw up splicing and so that female exon wont be expressed to check if work: Other thing they can do: instead of just making cut - can also supply a construct that has homology to place where cut and in that case instead of freaking out and going ends in NHEJ- cell will use clip and make place in genome to add strip — added in construct w GFP (put it in bc it glows) and it can be monitored by PCR — can then screen them and see whether they have long piece (mutant- added in region) or short piece- WT — or may be heterozygote- has one band at top of Gell and one at bottom (mifdle row in c)— did this to show successful in making these mosquitos Female dsxF -/- mosquitos also had other sex development phenotypes and were unable to feed on mice Then Q is — does this mutation In double sex gene actually affect fertility? WT make about 129 prodigy Male hetero and male make about same Male double mutant- ALSO is fertile!!— so carry mutation but able to pass on that gene But HOMO mutant female- don't develop properly bc missing double sex protein — couldn't mate and make offspring also a bunch couldn't get to that pt bc other developmental probs

started ( some beginning of notes) skipped to lecture 28 (NEED TO GO BACK TO 23 and begin again clarifications on crossovers and recombination and segregation during meiosis

chromosmomes inherited by certain individual are different than ones they will then put into gamete because they will have a combo of maternally inherited and paternally inherited alleles on one molecule-- you get a copy from mom and copy from dad and during meiosis 1-- those copies mix up -- so copy passed onto your offspring is mixture of the 2 **not every crossover results in recombination (NHEJ!) -- sister chromatids are actually not identical bc of recombombo! Meiosis destroys exact copy pass on- so when gets to meiosis 2 sister chromatids are no longer identitical

VIDEO- Molecular MARKERS- GO BACK TO NOTES FROM MODULES W PICS SSLPs

**Can use SSLP (microsatellites) or SNP and may converge on same gen of interest- can do the chunking tracking AB method and the A and B can be SSLPs or SNPs!! ***Genotyping but not sequencing but SNPs kind of bc sequence at that location SSLP- know size 2 individ have certain length of SLP and trait - then molecular marker linked to trait - bc both inherited this length- the mutation in gene that is causing mutation may be causing that phenotype SSLPs: decide by PCR genome has regions of extended repeats repeats can grow or shrink due to errors of DNA replication repeats can grow or shrink due to recombination -- they are homologous to eachother so could be that one version gets more repeats after the crossing over and other loses BC of reasons above: these can shrink and grow over generations so rather than A or B-- can be many versions of this region (lengths) in pop use homology in the primers of these sequences (so may have 3 green blocks in SSLp or 8-- but bf that- have homology in sequence- so create primers for either end -- then after primer polymerase will build and will create corresponiding length fragment ) then put individuals on gel individualizing 1: is heterozygous (has 2 strands of diff lengths one has 4 copies of repeat other has 8) individual 3: homozygous for 6 repeat bc only one band shown -- so how far down indicates length of strip and number of bands = hetero vs homo

guest lecturer

**Molecular charecterization- a historical approach sicke cell anemia - hemoglobin S and hemoglobin a mutation affects an endonuclease recognition site- site is affected by mutation in sickle cell CVn2 1- CCTNAGG Beta a - will be cute at that site (CCTGAGGAG) - pro-glu-glu BUT beta s globin- CCTGTGGAG (get valine at second positions- this nucleotide sequence not cut by Cvn1) So if do PCR—will get cut in beta a (look at base breakups) but then in Bs- no cut Then run PCR on gel and stain it Only 1 mutation in sickle cell- makes it simple Beta thalassemia Synthesis of beta globin chain is way down compared to alpga chain- they have hemoglymic anemia - get iron overload and cardiac malfunction Therapy for sickle cella nd Beth thalassemia Comes from fact that if u don't have any hemo A and have only hemo F (fetal)— you can be perf find w fetal hemo But normally at brith there is switch from fetal- get rid of it and then increase fetal A- which in sickle- the A is bad SO fine to increase fetal hemoglobin in adult life- been recently found that there is a gene, when turned on strongly inhibits fetal hemoglobin- so treatment— block the Bc111 a synthesis to increase fetal hemoglobin!! Bc111a inhibits fetal Hemoglobin Take stem cells from individuals w sickle- inowkc out the enhances for Bc111a - then put those cells into human and all have massive decrease in Bc111 synthshess and increase in fetal hemoglobin Array Every 1 mega base diff clone put down on slide 2464 clones Each printed in triplicate Put reference DNA on slide Hybridize ref dna w test dna coming in If image at that spot is yellow- then green probe coming in and red ref DNA have mixed— this is good If red- prolly deletion in test DNA bc didn't bind there If green- would have duplication— bc means many sequences trying to bind to that template Normal dots— along middle line- dots above middle line- duplicated regions (too much of green) **so could do SNP w 2 seq that should be all yellow- bc each person should have each version but if one green- missing piece- maybe lost in recombination or if extra red-- got 2 copies of that CNV!!! - 1 carom gave to another!!

microevolution

**muta Microevolution: describes changes in a populations gene pool from generation to generation rooted in 2 phenomena: either mutation brings in variation: BUT new mutations are relatively rare so they do not serve as vital source of genetic variation or major factor in promoting widespread changes in a population Random mutations within preexisting genes introduce new alleles into populations but at a very low rate These can be beneficial, neutral, or deleterious For new alleles to rise to a significant percentage in a population- evolutionary mechanisms (natural sleection, genetic drift, migration MUST operate— this not big enough change!) Mechanisms that alter existing genetic variation Natural selection Phenomenon in which certain phenotypes have greater reproductive success compared to other phenotypes — related to survival of members to a reproductive age Genetic drift Change in genetic variation from generation to generation due to random fluctuations. Allele frequencies may change as a matter of change from one generation to the next - tends to have a greater effect in a small population Which of the following phenomena is responsible for introducing new genetic variation in a population? Mutation **mutation always responsible for the new variation drift has big impact but not actually changing alleles!

sources of sequences to study evolution - benefits and drawbacks of each to answer evolutionary Q you are interested in - certain sequences better than others -inherited from both parents? inherited from more than one grandparent? found in every person in pop? undergoes recombination during meiosis? can develop new mutations? contains genes? contains intergenic non coding regions? ASNWER FOR ALL TYPES In considering any of the above sources of DNA sequence, at what locations would you expect the most mutations to accumulate in a given time period? Rank the following types of sequence from fewest mutations to most mutations and note any assumptions you're making.

- autosomes (22 of them) -sex chromosomes (X and Y) -mitochondria To place events on tree of human divergence Bc mito is so short- # nucleoitdes smaller - aren't enough sites to get change every generation even tho rate of change is higher from gen to gen end up w less overall mutations because just less genes to begin with Pros of Y chromosome Short Lineage is linear- just on paternal line in terms of inheritance Mutates faster than mito bc less genes Rare recombination Mitochondria sequences of humans Denisovans - wont be able to get Humansand deniso Cant see interbreeding using mito DNA bw bc mito cant undergo recombination so cant track changes but can use mito to show they all have common ancestor! but every samp,sample you have will have mito DNA in it- not every bone u dig up has Y bc not all boys so this is benefit of mito DNA CLASS NOTES: y kind pedigree- linear- traits will only appear in males mitochondrial inheritance: only passed on to offppring if mother carries it-- and if mother carries it- passes on to all of her offspring (BUT her boys won't then pass it on) mitochondrial DNA -not inherited from both parents: just mom -inherited from one grandparent: grandma -found in every person -does NOT undergo recomb - can develop new mutations -contains genes - contains intergenic / non coding regions Y chromosomes -inherited just from dad -inherited just from gradnpa -does not undergo recombination (X and Y don't combine) -can develop mutations contains genes/ intergenic non coding regions X chromosomes - inherited from both parents (only in XX) - inherited from both grandparents (if XX) -found in every individual in pop -undergoes recombination in females (XX- rarely in XY) -can develop mutations -contains genes - contains intergenic / non coding Autosomes -yes to all! ***meaningful conclusions: Y- will only show up in males AND only passed by males mito- will show up in males and females but only passed by females if type of DNA that doesn't undergo recombination-- cant see if interbreeding -- cant track chunks through evo tree-- can see share common ancestor Q2: would expect most mutations to accumulate at given time in sequences that are not used to encode proteins/ mrna bc those can mutate all they want and phenotype not changing functional RNAs< Exons in protein encoding regions< structural< introns and intergenic sequences (promoters and enhancers are a bit more stable- bc help in process but not really intergenic bc considered part of gene!)

melanin and regulating sequence exp

2 types of melanin eumelanin pheomelanin if have more eumelanin- darker skin 1st think did- looked at gene of interest If decrease MSFD12 (downregulating w shRNA (sh don't be exrpessed) increases eu melanin No MFSD12 rna at all when silenced by shRNA (binds to mrna and Mrna is degraded for MFSD12 so none expressed) **higher melanin if darker skin bc need pigment to protect u from stronger sun!- if lighter skin- need to absorb more and reflect less bc around less sun! 2nd- not linking at actual gene - look at promoter - saw that the sequence of this promoter varies and causes diff phenotypes so asked is this sequence difference controlling (or regulating) the MFSD12 (the candidate gene)— version of sequence associated w lighter and version w darker— is this enhancer controlling the MFSD12 gene- bc varies and by the gene tested in mice and zebra fish: ---* also bc didn't want to have to go take spectrums of how much eumelanin the gene made when promter on or off (less clear cut) INSTEAD - put a luciferase in place of MFSD12 (more clear cut, glow or not glow!) higher luciferase expression in lighter skin individuals - enhancer in front of luciferase (probably MFSD12 goes up)— if sequence makes luciferase go up it prob makes its actual target in chrome (MFSD12 go up!) ALSO MOre support also that MFSD12 is gene involved

SNP mapping vs whole genome sequencing!! (WGS)

Allele just means diff sequence at diff sites— it is expanding in use this word Will have diff combo at diff sites Each polymorphism has at least 2 diff at each site— we call A and B Each strip abobe is molecular marker— at any given site any copy of chromosome can be A or B Hetrozygosity is common THEN SCAN: can look at B allele frequency data- can from O (no B alleles at these sites (AA) to one where you have all B alelles (at top of scan) then dots in middle of scan are for heterozygotes (around frequency of .5)— bc half of it is B , AB, BA Doesn't exactly hit 50 but close -- so y axis= frequency of B allele- homo bb dots at top homo aa at bottom and heaters in middle what can tell from scan?? Heterozygosity common SNPs: single nucleotide polymorphisms - tiny diff bw 2 sequences - each one can be called an allele not necessarily in genes !!! can have nucleotide differences in regions of DNA that don't encode proteins mostly not associated with any phenotypic difference - in order to be useful- must be a common polymorphism (you don't want a sequence that is highly mutated, changing all time- want it to be common in pop in diff variations- this beneficial to seeing differences in individuals) *also- everyone in pop can't be homogenous at this site- bc then not useful- not polymorphic enough think about human chromosome: many different sites with known polymorphisms (on each chromosome) -- at any given site there can be n A type or a B type at that given site (the A version may have a C and the B version may have G allele) -- then next site can also have A or B form so along chromosome can read snip like AABABAA (so now just variation bw type A or type B- rather than which actual allele is there) *then each individual we know has 2 copies of each chromosome so could be homo AA w same type on each chromosome at that locus or heterozygote AB type SO SNP MAPPING which type of SNP pattern does an individual have in any region of the genome Whole Genome sequencing in contrast: sequences every single base pair includes known mapped SNPs includes new SNPs (there may be regions of the DNA that usually have 2 common forms of an allele that we don't know yet but bc sequence whole genome- have these new SNPs) includes variation of other types can identify previously unknown mutations (snips only show tiny pieces of sequences- SNP is not good way to find mutation itself - just narrow down places where mutation may have occurred

HWE mean fitness and altering eq and chi eq

Chi square test: used to evaluate the agreement between observed and expected data Can use chi square test to determine whether populations exhibits HWE for a particular gene Because the gene exists in 2 alleles- degree of freedom= 1 When geneticists have investigated other genes in various populations- a high chi square value is sometimes obtained and the hypothesis that the allele and genotype frequencies are in HWE is rejected (SO high chi sq value= reject null) In these cases population = in disequilibrium Ff The deviation indicates evolutionary change — if discover this then will go to see which factor is causing the change **HW biggest use is to understand changes in equal when they do happen and why- which type of selection! gene pool= only genes that were candidates in pop that made kids In 2012, there were 112 cheetahs in this population. During that year, 60 of those animals reproduced and 48 new cubs were born.-- gene pool is from 60 that reproduced bc other 40 not involved

molecular markers can be linked to an individual or lineage

Each person has a combination of alleles at different sites based on SNPs, RFLPs, SSLPs Recombination during meiosis can separate them into blocks **LOOK AT WS - have 2 copies of chromosome w diff versions of polymorphisms (either A and B at certain sites)- then across (from one version of chromosome to other- may be homo or heterozygous)-- then crossing over occurs and if have a polymorphism (version A or B) right next to another (right on top of diff A or B)-- Blocks of close polymorphisms will be maintained through several generations An individual with a series of the same alleles as another individual may be related -- w each generation boxes get smaller- so can see from generation to generation based on box commonality how closely related Then when redraw can note where blocks are— specific alleles still in parental order but 2 diff blocks in recombinant have now recombined SO, if see mutliple indiivusals in pop w same blocks together- could be idea that they are related- they inherited same blocks THIS is how 23 and me does it !!

selecting sequences foe evo studies

For recent sequence changes, select a relatively fast-evolving sequence For ancient events, select a sequence without recombination so that new mutations are the only source of changes Main idea: of looking for recent evolution- then would need fast evolving sequence relatively or else you wouldn't be able to see any differences Examine counter argument: if had sequence like histone H3— this shows very few sequence changes in the exons- if low rate of mutation and looking for something that happened very recently then prob wont see a lot of variation in that region BUT if want to look at ancient events don't want to select sequence that has recombination bc recombination complicates things — want one that doesn't undergo a lot of change- if looking for old events have to consider possibility that at any site it may change and then change back -- if undergoes recombination- could change and change back and move around a lot so for long term may see A in old ancestor and A now- but missing fact it changes so many time

race and genetics

For skin color- we know some genes and specific alleles are known to affect skin color Ways we think of race is not all captured in DNA sequencing- arg of article - some parts of our history that we think of as race that are not genetic aspects Reichs Arguments 1942: Race is a social construct with no basis in genetics. Supported by work in the 1970s based on protein expression showing 85% variation within races (West Eurasians, Africans, East Asians, South Asians, Native Americans, Oceanians, and Australians) and only 15% variation across races. -- genetics and protein expression show more diversity (85) from idivi to divi vs pop to pop Orthodoxy: racial differences are so trivial that they should be totally ignored fear of studying differences among populations - he builds this arg to bring it down he thinks- bc we have had already seen some (skin color) - we will see more so need to be prepared "it is important, even urgent, that we develop a candid and scientifically up-to-date way of discussing any such differences, instead of sticking our heads in the sand and being caught unprepared when they are found" He says:thinks white supremesists will do sequencing and publish paper and will cause social problems Critique of article- says a lot od fdiff things Says it is a social construct, not strictly genetic, but then says it is genetic, then points out that we need to be candid and thoughtful about how talk genetics Reichs research: in prostate cancer African-Americans have higher incidence of prostate cancer. Hypothesis: This is due to sequences specific to African lineages His research - higher incidence of prostate cancer in black men than in non black men in America (sequences specific that are causing the cancer and these sequences are specific to africans) Use SNPs shown to vary between Northern European and West African lineages. Look for associations between these SNPs and prostate cancer in "Admixed" populations of African-Americans Mixing of European and African lineages in the US happened within 15 generations, so blocks of sequence still exist. The size of blocks that are inherited together shrinks over multiple generations using self identified race to identify important sequences Use SNPs shown to vary between Northern European and West African lineages. Look for associations between these SNPs and prostate cancer in "Admixed" populations of African-Americans Mixing of European and African lineages in the US happened within 15 generations, so blocks of sequence still exist. Use a single marker to represent a large segment ~1Mb Survey a population at many sites throughout the genome, look for association between the version of the SNP prominent in African populations with the trait (prostate cancer) Look for SNPs shared across genome- then look for variation in these SNPs across the groups Peak at top is non random association with the trait Then smaller peaks arre not strongly associated They see that multiple SNPS have a signal on that y axis- one below top is a zoom in — see multiple markers that associate w the trait— all those peaks Then they go and look at the genes that are present in that area One of them is Myc= a known oncogene — known to cause cancer, increased cell division What he is doing is presenting method for identifying cancer causing genes **so saying race is a crude way of categorizing people but it works!! Populations were white people in Iceland, Europe, and the US. Trait is prostate cancer diagnosis Identified an SSLP on chromosome 8 that is associated with prostate cancer Confirmed the association in a population of African-American prostate cancer patients Did similar trial but on all white population Then found other polymorphism in other population including African American population and African American population tHIS shows that we didnt need African American population to understand the linkage of this gene**** reichs argument and intentions Using race to identify genes important in prostate cancer worked It can work for other human traits such as academic performance, family planning decisions, "intelligence" Acknowledge differences among populations, but don't discriminate against people because of them How not to talk about race and genetics: -- things are the way they are no need to put labels human and men differ bc have diff chrom sickle cell- not exclusively allele found in Africa- just wherever malaria present! Men and women have different genes and we don't ignore this to diferentiatie bw the 2 groups Sickle cell- not exclusively in africa- in fact it is just everywhere where malaria is bc malaria strong source of selection for heterozygosity for the hemoglobin genes Kidney disease- genotype/alleles— was thought to be derived from black people but now we know the genes- shows overlap bw medical treatment consequences and cstegeorizng by race— saying ppl cant be kidney donors as a whole bc African but actually jus took at genome for this gene and see if have or not bc all blacks prob don't!!! Excluded: biracial people , social scientists who know what these words actually mean and know about social and historical implications, native americans or asian people, non americans (or people who are from africa but live here? What do we call them?) What should be the role of self reported race in research or medicine? Bc race is a thing in America it is useful in medicine bc docs can be aware that broadly diff races are subject to diff stressors in our society or have broadly diff susceptibilities — this could be helpful but in a broad sense need to remember each individual so much variation How can we use things we know about other species to understand humans? These were inbred- so increased homozygosity- for humans we have a lot of heterozygostity so have to be careful when take something we observe in lab animals and apply it to diverse heterogenous mixed populations of humans

6.2 Extranuclear inheritance: mitochondria

Genetic material located inside the the organelle mitochondria in region known as nucleoid Genome is single circular chromosome- but nucleoid usually contains multiple copies of this chromosome Also- mitochondria also have often more than one nucleoid So multiple copies of circular chromosome in nucleoid and multiple nucleoid in one mito! # varies depending on type of cell and stage of development how many chromosomes copies of circular genome and how many nucleotides One mitochondrial nucleoid usually contains multiple copies of the mitochondrial chromosome Each copy of the mitochondrial chromosome consists of a circular DNA molecule that is only 17000 bp in length — less than 1% of a typical bacterial chromosome (SO SMLL) human mtDNA carries relatively few genes 13 genes encode polypeptides that function w the mitochondrion Carries genes that encode ribosomal RNA and tRNA (both needed for the synthesis of the 13 polypeptides that are encoded by the mtDNA) BUT mito need many more proteins to carry out oxidative phosphorylation and other mito functions— these proteins are encoded by genes w in the cell nucleus and then contain mitochondrial- targeting signals (so made in nucleus and then contain target which sends them to mitochondria to be used) In heterogamous species (species that make female and male gametes) mitochondria are usually inherited via egg cells Means the female parent passes mitochondrial genes to her offspring (bc think all dad really gives his his DNA strip— no organelles so no ability to give mito DNA) BUT this is not always the case

malaria and sickle cell genes, how detected and what gives what selection coefficient and CNV for globin family Why is it advantageous to have a family of globin genes?

HbA and HbS are 2 versions of the globin gene (we know there is a globin gene family so diff versions mad based on lengths of sequencing) 2 detect diff versions look at RFLPs cut by cvn1 HbA-(encodes glutamine) contains the sequence CCTNAGG cvn1 can cut this sequence so it is cut and 2 small fragments will show up on Gell of HbAHbA (both have been cut smaller) HbSHbS- (encodes valine) cvn1 can not cut so on gel- both big fragments chunk at top (bc big) HbAHbS (will have both patterns on gel) heterozygote advantage: explaining here why there is still a high frequency of recessive alleles that when homo recessive are deadly (heterozygote advantage explains why the allele does not go towards fixation of one of the 2 alleles!) ex: HbS allele of human b globin gene homozygous HbSHbs= sickle cell HbSHbS homozygous has lower fitness than HbAHbA BUT heterozygotes HbAHbS have higher level of fitness than either homo in areas where malaria is endemic!! HbSHbA- have higher chance of survival from malarial parasite so need the HbS allele to be maintained! SO an allele will not go to fixation of the heterozygote has higher fitness than either homo in a certain area!!- heterozygote add-balancing selection common misconception: natural selection ALWAYS eliminates weaker alleles balancing selection: heterozygote advantage for genetic variation involving a single gene (so just looking at diff alleles of 1 gene) balancing selection may arise when the heterozygote has a higher fitness than either corresponding homozygote equal is reached here: when both alleles are maintained in the pop if know relative fitness values for each of gentiles- allele frequencies can be calculated selection coefficient: measure degree to which genotype is selected against: s=1-w (1- fitness= selection coeffiticnet) if high fitness- not selected against so small selection coefficient genotype w highest fitness s= 0 w=1 pop reaches equal in pop undergoing balancing selection when SAAp= Saaq p and q equally selected against (so if both homos equally selected against- heterosexual may reign!) **CLARIFICATION - HbS and HbA are diff alleles that help make the hemoglobin protein hemo globin and myoglobin are the proteins all started off w ancestral globin (1 globin codes 1 protein to help bind oxygen) then- CNV differentiation of primordial myoglobin vs primordial hemoglobin (CNV creates chromosomes that now has 2 similar sequences (but more specialized) CNV again -- the hemoglobin gene undergoes CNV again (at site of CNV- gained another copy variant- further specialize) alpha vs beta chains ** the specific diff alleles looked at in sickle cell is HbA HbS THESE are 2 diff versions of the beta hemoglobin gene Homologous genes within a single species are called paralogs (what make up gene family) (2o 2 paralogs on same chrome) Globin genes encode polypeptides that are subunits of proteins that function in oxygen binding (globin is gene bc g and g)- ex: hemoglobin =protein found in red blood cells (function= carry oxygen throughout the body) The globin gene family is composed of 14 paralogs that were originally derived from a single ancestral globin gene!! Evo analysis of gene Ancestral globin gene first duplicated about 500 million years ago and became separate genes encoding myoglobin and the hemoglobin group of gens Primordial hemoglobin gene Duplicated into the alpha and beta chain gene !! (so essentially alpha and beta are 2 versions (slighlty diff) of gene that helps form hemoglobin!)— these than also duplicated to form genes located on chromosomes 16 and 11 Currently: 14 globin genes found on 3 diff human chromosomes !!! The accumulation of diff mutations in various family members has produced globulins that are more specialized in their function! ( more genes, more variation, more to select on, better function) Ex: ancestral globin (first and only gene) Duplicated into primordial myoglobin and primordial hemoglobin (both variants of same gene so both used for oxygen transport BUT myoglobin is better at binding and storing oxygen in muscle cells and hemoglobins (plural bc duplicated again) are better at binding and transporting oxygen via red blood cells) (mya just wanted to store her things whereas I wanted to run around- transport) ALSO Also diff forms of globin are expressed at diff times in life — alpha globin gene is turned on in gestation and it is not till after birth (beta) that beta is turned on — SHOW the expression of globulin genes reflects the diff in oxygen transport needs of humans during embryonic fetal and postpartum stages of life ***SO if have variation or mutation in beta- not creating a lot of necessary protein- could up expression of protein that is usually just gestational!

molecular clock

Molecular clock Assumes - if you have sequence with a constant mutation rate that is calibrated to samples you have (not too fast)- can then look at number of mutations or differences and infer time since divergence Had fossil samples- looked at diff sequences they could get from those- based on dating of fossils- could get sequences from them and identify mutation rates or changes of sequences throughout time If comparing human vs neanthrdal need to make sure they are evolving under same molecular clock to make meaningful assumptions

genetically engineering mosquitos - fighting malaria and zika w modified versions first trying to target plasmodium

Human induced evolution in a species relevant to human health and comfort Genetically engineer mosquitos- alter the way they carry malaria or interact w humans and may be able to change mosquito carry malaria to human 3 parts: Humans- get malaria Anopheles (mosquitos) carry disease bw humans Plasmodium- parasite itself that causes the disease So if trying to combat malaria- can address any of pieces of this puzzle could try to target parasite: Plasmodium malariae all cause malaria in humans. Other types of Plasmodium parasites cause the disease in animals. Some of these are also able to cause "zoonotic" malaria in humans too. These parasites have a complex life cycle. They are amplified in human livers and red blood cells. They also form gametes in human blood cells which come together when ingested by a mosquito. Parasites develop and replicate in the mosquito and infect a human via a mosquito bite plasmodium goes into liver enters liver cells replicates in high numbers and moves into blood where can proliferate All about damaging effect on red blood cells Other thing they can do is split into that bottom half- form gametes- do repro- Recombinate so pop continues to proliferat plasmodium goes into liver enters liver cells replicates in high numbers and moves into blood where can proliferate All about damaging effect on red blood cells Other thing they can do is split into that bottom half- form gametes- do repro- Recombinate so pop continues to proliferat Mosquitos play important role in spreading disease Vaccines designed to target Plasmodium Specific antigens that elicit human antibodies that target Plasmodium gametes can disrupt fertilization in the mosquito—limiting transmission of the parasite (doesn't really help the person with malaria ) Whole sporozoites injected intravenously elicit a strong immune response that blocks disease development Approaches -target plasmodium, target parasite itself — use specified proteins from plasmodium specific to gamete form to illicit response from human immune system - -- targeting the gamete form- Better way: Use whole live malaria parasites as vaccine!! —very effective- injecting small amount illicitness very strong immune repsonse that blocks proliferation there and leaves ppl immune to future

penetrance

If a gene is fully penetrant, the genotype determines phenotype Some genes are only partially-penetrant, meaning that individuals may have the allele that causes a trait, but do not have the trait due to environmental conditions due to undiscovered interacting genes due to our inability to accurately score or categorize the phenotype

natural selection vs drift

If dom allele not necessarily selected for If many sites go to fixation- consistent w bottleneck- favoring fixation at any given site WHEREAS if one allele rises to prominence at 1 site- seems more like selection over drift!! Selection vs drift Something in world favors one allele over other Drift- allel frequncy chnages but this is just do to drift- not optimal or better bc natural random occurrences

Evidence of directional selection of human pop- lecture 26!

LOOK AT PIC OF HAPLOTYPES BF AND AFTER SELECTION! rows represent diff individuals and squares in each row represent haplotypes (maybe SNPS, SSLPS etc) 2 versions of polymorphisms so squares either gray or black BUT- in one row (1 person) has red square single individual mutated (not poly) THEN after selection- at that chunk-- all in pop now have red this is selection bc happening at one place in genome- not all changing- and all got after so obvi being selected for - directional selection!! evidence of direct selection:; LOOK AT GRAPH y axis is gene diversity and line in europeans drops to 0 at that SNP- means everyone in pop has that version map of world: blue SNP- T associated w light skin yellow- A- dark (all dots in Africa and Eastern Europe) in Northern Europe: predominantly blue so allele is fixed at that location! means every individual in that pop has the same allele

Ancestry and Forensics - lecture 22

LOOK AT PIC OF SEQUENCES NEXT TO SNPS, RFLPs, SSLPs Each person has a combination of alleles at different sites based on SNPs, RFLPs, SSLPs Recombination during meiosis can separate them into blocks Blocks of close polymorphisms will be maintained through several generations pic shows: linkage map- genes or sequences that have been identified by traditional mapping -in bw yellow sequences, has distances bw them denoted then have Red blue and green chunks that indicate locus distribution **it is saying at this ;locus HOW MUCH OF THE DISTANCE IS DNA MARKERS (SSLPS RFLPS AND HOW MUCH IS GENES - red and blue are DNA markers and green is gene and chunks of green way smaller- so must of space of each locus (chunk) are DNA markers! earlier: linkage bw genes but genes such small part of genome that becomes challenging to get full picture so NOW Looking for linkage between molecular markers and each other or bw molecular markers and a given trait or set of traits scans of polymorphisms tell us Each polymorphism has at least 2 diff at each site— we call A and B Each strip abobe is molecular marker— at any given site any copy of chromosome can be A or B Hetrozygosity is common Then can look at B allele frequency data- can from O (no B alleles at these sites (AA) to one where you have all B alelles (at top of scan) then dots in middle of scan are for heterozygotes (around frequency of .5)— bc half of it is B , AB, BA Doesn't exactly hit 50 but close So homozygote at top and other homozygote at bottom - heterozygosity in middle— what can tell from scan?? Heterozygosity common commercial sequencing services can find your relatives - how 23 and me does Predicts likelihood of relative by finding blocks of molecular markers that are same w others Genetic ancestors vs geneilogy- genealogy is looking for also birth and marriage records, what was family doing back then vs genetics just genetics commercial sequencing services can find crime suspects - golden state killer Found bc DNA evidence found at crime scene Was sent and used to quarry known individuals then they found the suspect based For this— found him bc his 3rd cousin was in the data base- genetic makeup fairly similar and then authorities can track down how GED and trees: look at pic Bold- 2 individuals that were in the GED match database GED= can upload raw seqeuncing file and it has more info you can access Crime suspect matched to these 2 then use genealogy - reporting and investigating to build these family trees Found connection bw these families- there was a half sibling that connected the 2 identity inference of genomic data using long range familiar searches- created database of 1.8 million genome sequences 1. Create a database of 1.28 million genome sequences 2. Remove 1st cousins or closer (bs so well matched) 3. Look for blocks of matches that indicate identity by descent (30-600cM) 4. 60% of the time, they found a match- highest likelihood of matches for those with Northern European ancestry because most of the individuals in the database are of the ancestry Size of blocks inherited together shrinks over multiple generations!!!- this is how it works — bc more recombination- blocks keep recombine and keep breaking up blocks into smaller bloclks- as go blocks smaller and smaller SO can look at blocks of diff lengths to see how closely related —longer sequence, can find closer relative! Second counsin- will have at least 100 cm in common - found third cousin of killer procedure to do this 1- 1. Create a database of 1.28 million genome sequences 2. Remove 1st cousins or closer (bs so well matched) 3. Look for blocks of matches that indicate identity by descent (30-600cM) **once narrow down to range 30-600 - the 600 is largest block- closest relative (1st cousin once removed - then 2nd cousin then second cousin once removed then their cousin third cousin once removed all the way to smallest block= 4th cousin 1 removed (you only share that tiny little block 2 them bc of all recombination) SO- longest blocks = closest relation BUT- probability of attaining long blocks matches in system is low EXISITING DATABADES INCLUDE 3rd cousin matches for majority of people so to make more accurate- can filter start w all 4th cousin sequence matches for ex:- filter to genealogical match (birth, family records, what fam was doing back then) geography sex age (had to have been old enough for relation to work- or old enough to kill) ** the key is that it is the size of blocks that are inherited (TOGETHER) that shrinks over generations SO- if 2 ppl have relatively large block similar -- they are closer related than someone where that same box has split in 2 (would be one gen after!)-

Mitochondrial DNA

Mitochondira- used to say when did these samples diverge passed by mom- given to both To look at ancient events- use mitochondrial DNA-!!! inherited from maternal line - blue version of mitochondrial starting at blue circle and passed on Note not present on left 3rd row bc have diff mito bc have diff mito line HIGH mutation rate compared to autosomes but they are so short that it is likely that there will be 0 changes bw generations bc just not that many nucleotides to mutate AND too short for recombination so -== overall likelihood of no changes in a few generations --**overall likelihood of no changes in a few generations means that if DO see change- ancient change- stemming from common ancestor most likely!! One mitochondrial nucleoid usually contains multiple copies of the mitochondrial chromosome Mitochondrial form tubules in cell and are constantly fusing and separating from eachother- often around 100000 copies of mitochondrial genome per human cell Means that like to get quite a but of mitochondrial DNA from ancient remains- helpful for extracting DNA from old samples- easier than extracting autosomes SO EASY TO EXRACT- 100000 copies per cell!1

Questions A segment of DNA that is located at a specific site along a chromosome and that has properties that allow it to be uniquely identified using molecular tools is called a(n) The association of a molecular marker that is transmitted along with a genetic disease in a family pedigree indicates that the molecular marker Molecular markers that vary from individual to individual within a population are said to be Which statements describe (CA)n sequence ? Which type of molec marker does not vary in size bw individuals? If a polymorphic marker is close to disease causing allele then marker will

Molecular marker Molecular markers are useful because they are polymorphic, which means they vary from one individual to another within the same population. is likely to be close to the gene that causes the disease polymorphic Found in human genome Thousands of copies in genome Can have 5 to 50+ repeats ex of a micro satellite N indicated number of repeats SNP-- not looking at size of strands w this! Be linked to disease causing allele!!! This is whole point!! Distance bw 2 sattelites can be counted by counting frequency of crossing over bw them

Two types of factors can govern microevolution. On one hand, mutation is a constant source of new genetic variation. On the other hand, mechanisms, such as natural selection and genetic drift, can act to alter existing genetic variation. How do those two types of factors compare in terms of the magnitude of their effect on allele frequencies in a population?

Mutations have a negligible effect on allelic frequencies, while mechanisms, such as natural selection and drift, can have a dramatic effect. -- mutations have a negligible effect bc starting at one person!! that has to propagate into pop

VIDEO- Molecular MARKERS- GO BACK TO NOTES FROM MODULES W PICS RFLP and RFLP to detect malaria and sickle cell risk

RFLP: detected by PCR followed by treatment w restriction enzymes here- need restriction enzymes that cut double stranded DNA at specific sequences some polymorphisms (SNPs or insertions deletions can either create or destroy the sites in DNA that code for cut to be made 1- amplify specific site using PCR - attempt to cut it w an enzyme 2- asses fragment size on Gell can asses whether cut or not based on agarose size detected on agarose gel RFLP to detect malaria and sickle cell risk using beta globin gene 3 diff sites (LOOK AT PIC) on/around gene where CVn1 can cut Cvn1 cuts when CCTNAGG (CTT (anything)AGG) HbA allele (version of B globin gene) (particular sequence of nucleotides)- encodes glutamine GAGG (SO THE N in this case is A!!) -- the HbA allele is cut by Cvn1 bc has code required for cut!! HbS allele (other version of B globin gene) encodes diff protein= val bc has C TGTGG (diff ending than necessary for protein cut indication)- cant be cut by Cvn1 SO go to portion of Cvn1 gene that covers this Cvn1 site **so go to specific site in gene that we know either is cut or not cut depending on sequence (this is polymoprhism- cut v not cut!!) and amplify , look for diff in lengths to determine HbS presence vs HbA prescence RFLP to detect malaria and sickle cell gel: HbAHbA (A is A ok to cut!) will have sequences cut but Cvn1-- so will have fragment on gel cut into 2 smaller sequences HbSHbS - cut cant be made amplifies fragment of beta globin gene- 1 strip at top including whole gene (bc gene not cut into smaller fragments) HbAHbS will have both amplified long version at top and 2 smaller ones farther down

random genetic drift and bottleneck effect

Refers to changes in allele frequencies in a population due to random fluctuations As a matter of chance! rate of genetic drift depends on :depends on pop size and initial all frequencies (what was prvelanace of allele to begin w?) once become monomorphic — does not fluctuate any further **what did when taking out peas- took some out of pop - pop left over less diverse- eventually one goes to fixation BUT ONLY IF POP SMALL WILL RESULT IN FIXATION- EVEN IF NOT BENEFICIAL IF POP IS SMALL In large pop- allele frequencies over time fluctuate much less bc random sampling error (Sampling error occurs when researchers take a random sample instead of observing every individual subject that comprises a population.) is expected to have a smaller effect — the genetic drift will still lead to allele fixation in large pops just takes longer for effect to occur! **line for generations on x anxious vs frequency of A on Y— for the n=20 lines way steeper and for N= 100 way more gradual almost flat lines Process is random w regard to particular alleles How many new mutations do we expect in a natural population? Avg number of new mutations depends on mutation rate (mew) and on number of individuals in pop (N) — if each individual has 2 copies of gene of interest- expected number of new mutations in gene is Expected number of new mutations= 2N(mew) This eq shows!!: New mutation more likely to occur in larger pop (on pop sides of eq so direct)— makes sense bc larger pop has more copies of genes to be mutated How likely is it that any new mutation will be either fixed or eliminated from a population due to genetic drift? Probability of fixation= 1/2N Saying that prob of fixation is same as the initial allele frequency in the population The higher N is, the lower the chance that a new mutation will become fixed. Prob of elimination 1- prob of fixation (bc opps) 1- 1/2N N has opposing effects with regard to new mutations and their eventual fixation in a population N very large- new mutations are much more likely to occur BUT each mutation then has a greater chance of being eliminated from the pop due to genetic drift (bc slim is 1- fixation and if N is large fixation ins small (in denim!!) N very small- prob of new mutation is also small but if they occur- likelihood of fixation is relatively large!! If fixation of a new allele does occur, how many generations is It likely to take? Form also depends on number of individuals in pop t(with hat)= 4N t= avg number of generations to achieve fixation N= number of inidivuslas in the pop (assumin males and females contribute equally to each succeeding generation) Obvi from this- as N increaases- allele fixation takes longer The larger the population, the more generations will be required, on average, to achieve fixation of a new allele. Bottleneck effect Pop reduced dramatically in size by events like earthquakes or floods, droughts or human destruction of habitat ( so randomly erasing alleles-not choosing like natural selection) Randomly eliminates w out regard to genetic composition Bottleneck effect may initiate genetic drift bc the pop of survivors may have all frequencies that differ from those of original population ALSO allele frequencies are expected to drift substantially during generations when the pop size is small (which bottleneck is doing!) In extreme cases- may even eliminate alleles Eventually pop that experiences bottleneck may regain size but new pop will have LESS genetic variation than original large population-- think hourglass

common types of molecular markers and sequence tagged site definition (STS)

Restriction fragment length polymorphism (RFLP) Site in genome where distance between 2 restriction sites varies among diff individuals- these sites are identified by restriction enzyme digestion of chromosomal DNA and identification of DNA fragments that vary in length **region that differs in size when cut w restriction enzyme Amplified Restriction fragment length polymorphism (RFLP) Same as an RFLP except that the fragment is amplifies via PCR intstead of isolating the chromosomal DNA ** region that differs in size when amplifies w PCR Microsatellite Site in genome that contains many short sequences that are repeated many times in a row- total length usually 50-200 bp and length of given micro satellite may be polymorphic w in a pop (H vs h have diff lengths both common in pop)— they are isolated via PCR— also called short tandem repeats (STRs) and simple sequence repeats (SSRs) Single nucleotide polymorphism (SNP) Site in genome where a single nucleotide is polymorphic among diff individuals- these sites occur commonly in all genomes and theya re becoming more widely used in mapping of disease causing alleles and of genes that contribute to quantitative traits that are valuable in agriculture Site where one nucleotide polymorphic

Ecological examples of the impact of genetic drift: bottleneck

Temporary reduction in population size Likely to reduce diversity Population may return to pre-bottleneck size, but not pre-bottleneck diversity -- think of hourglass demonstration (top part of glass may have blue green and orange balls, then through thin part only blue and green go- now once pop grows again only ble and green can be present!) New Zealand black robin:The 1 female was able to successfully mate w one male- researchers would steal their eggs and put in diff place so they would think needed to lay again—to create more and more species - every robin descended from same pair

What if this gene drive is actually bad? Will all mosquitos have it forever?

The Cas9 protein is sometimes put under the control of a promoter that is sensitive to the amount of a modified amino acid in the environment. In the absence of this version of Lysine (BOH), the Cas9 protein will not be expressed and the CRISPR system won't work. The construct will then just be inherited normally. Some CRISPR delivery vectors are relatively short-lived in cells and persist long enough to allow gene editing but then are degraded. -- so may be degraded by time mosquito goes to have second round of babies

Darwinian fitness directional selection

The relative likelihood that one genotype will contribute to the gene pool of the next generation rather than other genotypes (chosen over other genotypes) Can asian a fitness value to each genotype class based on the likelihood of individuals with that genotype surviving to reproductive age ex: for every 5 AA individuals that survive, four Aa individuals survive, and one aa individual survives By convention- genotype with the highest reproductive ability is given a fitness value of 1 Directional selection Favors individuals at one extreme of a phenotypic distribution that are more likely to survive and reproduce in a particular environment Diff phenomena initiate the process of directional selection New allele may be introduced into a population by mutation and the new allele may promote a higher fitness in individuals that carry it —- if homozygote carrying the favored allele has the highest fitness value, directional selection may cause this favored allele to eventually become the predominant allele in the population — may even give rise to a monomorphic gene so to adjust for the fact that pop is evolving you * by the fitness of each allele If pop is changing due to natural selection—- these may NOT add up to 1 as in HWE Instead add up to mean fitness of the population (w- w hat on it) If change in population has occured bc AA for example has highest fitness than the allele frequencies of A will increase and a will decrease Interesting feature of natural selection!!: increases the mean fitness of the population What are the consequences of natural selection at the population level? After one generation the population is better adapted to its environment than it was before in the preceding generation — new pop has greater reproductive potential than the preceding one — if keep doing calculation for next generations mean will keep increasing and so will A as a decreases Eventually directional selection may lead to fixation of a beneficial allele BUT new beneficial allele is in a precarious situation when its frequency is low (BC— genetic drift is likely to eliminate new mutations, even beneficial ones, as a result of chance fluctuations) Power of genetic selection can be seen w insect like mosquito resistance to pesticides

In this case- researchers found 4 genes involved in skin color. What would be next step you would take?

This is all forward genetics: start w phenotype of interest - find gene (or several genes) In this case- researchers found 4 genes involved in skin color. What would be next step you would take? Complementation test- tells u whether 2 diff mutations are in same gene ( we already now that they are bc mapping to diff places in genome so we already know this!) *would perform these bc have identified 4 genes that may relate to skin type - now want to see if maybe 2 of those genes are interacting to creat the skin type in complementation relationship def: phenomenon in which 2 diff parents that express the same or similar recessive phenotypes produce offspring w wild type phenotype Typically occurs bc the recessive phenotype in the parents is due to homozygosity at 2 diff genes — one parent was CCpp and other ccPP SO, in F1 gen, the C and P alleles which are WT compliment the c and p alleles (all hetro now) **complmentations tests ALLOW: you to see how many genes responsible for traitr- in pathway (make complementation groups- if both 4 and 3 cant make arg when on gel then must be mutated on same gene pathway- same complex group) BUT 1 and 2- they complement eachother bc they are mutant at diff spots so the WT of other cancels out — also cant mate too humans together! This is humans don't do this Epistasis analysis- could look at existing pedigrees w in these populations - but w in family probably have same SNPs — also would need to cross- so don't do this epistasis looks at : when the genotype at one locus determines the phenotype regardless of the genotype at a second locus. Partial credit for partial answers "one gene controls another gene" Suppressor screen- not in humans BIG THING when one mutation reverses the effect of another mutation, causing the double mutant to have a wild-type phenotype.**NOTE: in suppression double mutant has WT-- in epistasis double mutant what have to do for humans: "candidate gene" approach Study related (orthologous) genes in a model system to learn about the encoded protein Alter the expression level of related genes in a model system and look for phenotypes Express this gene in a model system and look for phenotypes Start w gene - what we identified using SNP array was a bunch of genes Orthologous- not exact match but similar function Can alter expression level of various genes in model system

study that capitalizes on getting mitochondrial DNA ** and what other piece of genetic material confirmed this?

Worldwide human mitochondrial haplogroup distribution from urban sewage Taking sewer samples in order to do surverys of microbial pathogens in diff regions (bacteria mostly) in diff places If can survey location can predict disease outbreaks based on this info One thing they noticed- a lot of human DNA in these samples- could get whole mitochondrial genomes from these samples as well- could get enough mito DNA to estimate population composition — could try to get a picture of white types of people living in which communities Early mt genotypes were done— phylogeny of mitochondria haplotypes shows how mitochondrial sequences are related to one another and connected to common ancestors Found that there was variation in mutation rate at diff parts of mitochondrial genome!! Part of mitochondrial genome that evolved at diff rates in diff populations!! If evolving at diff rates means that you cant use that as a clock (if one persons clock moving faster than others cant use that as standwar) So in this type of tree- deep branches so more time of divergence since breaking from common ancestor- if shallow short branch- implies 2 sample very similar to eachother but then if huge line until nect CA- then lot of time bw those populations - have to make tree based on time of mutation rate then from diff pops (so diff pops alive at same time but mito evolving at diff rates) Short branches show less diversity in European lineages— bottleneck possibly due to migration- shrinking in population possibley due to migration or disease that could reduce the genetic diversity!! (making branches shorter) Have mitochondrial eve which separates humans from other chimps- every human has mito that came from one person 150000 years ago - common mitochondrial ancestor "eve" from 150000 years ago based on molecular clock estimates !! BUT our shoaled mitochondria is not from first human female— and this person wasn't alone- prob other mitochondrial sequences but progeny of 1 individual ended up dominating the population and that happened are 150000 years ago In this type of tree, deep branches show diversity e. g. two San sequences are quite similar, two Mbuti sequences are quite similar (short branches) but lots of diversity between San and Mbuti (bc both long lines away from common ancestor!! These short branches show less diversity in European lineages => bottleneck possibly due to migration Early mt genotypes were done with RFLPs This is the first one to use the entire mt genome sequence*** so they were RFLPs done to relate haplotypes but this study- whole mito genome Phylogeny of mitochondria haplotypes shows how mitochondrial sequences are related to one another and connected to common ancestors. Common mitochondrial ancestor "Eve" from ~150,000 years ago based on molecular clock estimates each letter represents haplotype mito allowed for this- could have done same thing w Y or whole genome but mito makes this linear other piece of DNA that confirmed single common ancestor- Y chromosomes also point to a single common male ancestor from ~250,000 years ago

Tuco Tuco ex- look at PWP graph and what is rate of fixation determined by

allele freq!! (rem always allele) tracking pop and notice allele freq goes to 0- is it evolving? Yes- directional - if 1 allele fixed at 0-- would assume other horizontal line for fixation of other allele at one if around .5 (heterozygotes- balancing selection- hetero adv) bean activity: selecting tuco tuco black or white - randomly, color= trait that does not affect their fitness, not under natural selection what did- selected 4 random from cup that will contribute to next gen- then duplicated them (so maybe selected 3 B and 1 W) then duplicates so had 6 B and 2 W- still pop=8 repeat until get then repeat- once you pull 4 B- will duplicate to *B and remain black for remainder At end: most popular was 0 W 8 B or 8 B and 0 W How many generations did it take to get this 0-4 generation were groups who reached fixation early 5-12 fixation middle rate 13-17+ still working on fixation When asked what was going to happen ahead of time we thought frequencies wouldn't change bc lack of selection— but actually shows we do!! Many of our populations reached fixation SO, even tho we see one allele going to fixation- could be random process doesn't mean selected for **reaching fixation but not because of selection reasons!! THEN- repeated procedure w 2 W and 2 B- got to fixation quicker Conlusion!!: smaller pop- more susceptible to drift Stories we tell ourselves abt evolution is that If a trait is prominent doesn't mean its good- better than others or bc trait low doesn't mean worse— this is WHY in humans we see a lot of polymrphisms- one is not mutant and one is not normal they are both fine pop size and original gene frequency are determinants of rate of fixation -- if fixation is coming from bottleneck - already that gene was present in pop so many can help to pass it on but if fixation is coming from mutation-- only 1 indiv that is beginning the fixation!!- longer graph where lines start at .5- sharper lines bc can reach fixation faster bc have some alleles frequency already when start at 0- lines towards fixation more gradual

Evolution: selecting sequences for evolutionary studies- 28 predicting ancestor sequences which of following kinds of sequences would be least helpful to resolve the confusion about the ancestor sequences? outgroup more sequences from w in the pop to build a stronger consensus about A/T sites fossil sequences from related organisms from time proposed species splits

diff types of sequences that we are interested in when looking at long term vs short term evolution can use current sequences to predict ancestor sequences: if have 5 diff sequences and all have certain spots where same - ex: say have G at locus 5 on all of them- then can confirm that G must be there and that must be from ancestor sequence if comparing sequences and looking at relations- get rid of parts where all same- then look at remaining parts- who is most similar line up sequences--then notice- are there differences in sequence length- if so- then find deletion location in shorter sequence (5) shorter sequence aligns perfectly with 4 if insert space where deletion is now- 2 hypothesis: sequence 5 had deletion, others had insertion **want lowest amount of genetic events= most probable so seq 5- deletion compare each strand and make assumptions about ancestry-- look at 2 strands and see where they differ- lets say differ at A and 1 has A and 2 has G- if all others have A there- A must be from ancestral sequence! answer to Q: outgroup sequences that are less related to these populations- THIS IS NOT CORRECT- outgrips are super helpful!!- they can help inform you about the common ancestor0— for ex looking at plstypus- look at something that is not mammal then look at outgrip vs mammalian sequences- gives more info about where platypus lies in relation - fossil sequences from related organisms from time proposed species splits If not actually sequencing someone from node (common ancestor) sequencing someone along branch point- may be diff than CA so not actually telling you anything But if could find fossil that is CA and sequence it- this would be helpful!! - more sequences from within the pop to build a stronger consensus about A/T sites and other If you pick more sequences from w in pop likely getting more examples of same sequence- more ex of either same sequences or other ways to come from those same individuals Choosing more related individuals likely to pick up close genetic relative so might just be getting more numbers of same data- not getting anything new from diff parts of tree- Consensus is not same thing as prediction of ancestor!- just bc everyone has G there now doesn't mean thats what Common ancestor had!! LEAST USEFUL

in considering exons- some sites under more selection than others- based on genetic code- which sites most likely to be conserved

genetic code-- groups encoding same protein are grouped bc share same first and second nucleotide-- but third can differ 3rd- nucleotide - not subject to selection bc change would not result in phenotypic difference ***If something is subject to selection- changing it would mean changing phenotype- so imp parts of DNA will not be subject to selection AND If not subject to selection- than can mutate as much as need*** SO If you were interested in investigating recent evolution in a population, what types of sequences would you focus on (considering #1 and #2 above) and why? interested in recent evolution look at intergenic sequences, introns etc old evo- other seq!

Hardy implications if you have better estimate of certain pop makeup ex w sickle, malaria

if we have more info about a genotype, we can change HWE to better estimate - if know more about pop- change HWE to better estimate it balancing selection- heterozygote advantage in certain parts of world- all concentrations will be diff based on certain things in general w Hbs vs HbA - HbS (q) value would be smaller bc homo HbsHbS have sickle- but need allele for heterozygous advantage in malaria resistance in africa- parts where need it o we see observed frequencies of the heterozygote are higher than expected by the HWE in these places bc of the heterozygote advantage in these places ** observed hetero freq is higher than expected by eq For trait like this— selection against HbSHbS phenotype in form of disease but selection that favors HbSHbA genotype= malaria resistance To correct for this: adjust hardy to consider variable fitness of diff phenotypes fitness = reproductive success fitness w is relative compared to other genotypes If looking at balancing selection- use selection coefficient- 1- fitness= magnitude of section on diff phenotypes Know that when selection occurs- whether directional or balancing that we have to modify predictions in order to explain what is happening in real population for rest of pop: not African -- European for ex: look at graph: at specific allele (distance in kilo base pairs on bottom) and gene diversity on Y-- europeans plummet to o in the diversity at and around this gene bc no selective advantage to carrying the HbS (not by malaria) so all have HbAHbA-- whereas africans have genetic diversity at that locus!

VIDEO- Molecular MARKERS- GO BACK TO NOTES FROM MODULES W PICS SNP

main types : SNPs, RFLPs, SSLPs (microsattelites) the DNA sequence here is the phenotype (ex short or long- is phenotype needed to predict likelihood of linked alleles) SNP: detected by sequencing, hybridization or PCR w stringent primers -detected by micro array: you probe sequences attached to a solid surface (chip) - shown 2 chips w covalently bonded sequences ( several copies of 1 sequence on 1 chip and several copies of other on other chip) -THEN: pool of sequences from sample of interest (want to see if these sequences differ from eachother ) - 1 sequence coming in. may base pair w one chip whereas diff sequence (diff only in 1 nucleotide bc SNP- will match to on bottom chip) **if sequence comes in that does not match to one probe (think one prob may be A and other B polymoprhism- if other seq has diff- will float off)- bc just comparing these polymorphisms FROM HERE can tell how many copies of specific sequence in a given example and can look for which spot on chip labeled another way chip can be made **put bead on the chip with covalently attached probes- sample DNA comes in again and base pairs w sequences flying off of beads (probs) -- 1 bead will have sequence that base pairs to A version whereas other has sequences flying off that correspond to B version ways these detected: probe (flying off bead) is slightly longer than sample so then add dNTPs to extend each incoming strand and extend sample by single nucleotide that is detectable both chips: presence and intensity of label (not identify of what is added_ confirms that seq is present in sample - don't care abt actual nucleotide added- but if one chip has a bunch of needs lighting up -- we know a bunch of sequences have bound to reference sequence of version A lets say so we know version A is present a lot in sample SNPs detectable by sequencing: would have reference genome location of known polymorphism then can look for ratio of A to G for ex at that site can do whole genome sequencing then align reads that overlap at site of polymorphism and see what present

SNP explained

make 2 diff beads for 2 diff people then put their sequences (amplified by PCR) on the beads say A bead and B bead- if both ppl have more of their sequences attached to A bead- then we can say they have a commonality in this region- don't know exactly what bc based on intensity not actual gene but allows for honing in on region!! sequence of DNA has been designed based on what usually around particular SNPs **we have knowledge about known polymorphisms so if want to see what one person v another has-- put knowledge of info around diff located ones on chips fragment extends just up to polymorphic position but doesnt include it then incubate w incoming strands to base pair -- depending on length etc may pair to bead at diff positions SO sequence is hanging off by one longer nucleotide (that is at site of polymorphism)-- then have fluorescent tags that come in w 2 diff nucleotides it could bind to (one one color one another) If homozygous-- whole bead would be same color bc all C's lets say coming in to bind w all G 'g BUT if hetero- some of beed one color some of beed other color -- SO MISTURE OF 2 colors- bead would fluoresce yelloW!! have gone through genome and identified polymorphisms in SNPs genes SSLPs etc by saying ok at this location in most people it is either this or that-- then saying is the commonality

Linkage mapping and micro satellites LOOK AT PIC IN READING NOTES ASSOCIATED

makes refined map of genome-- researchers identify diff polymorphic sites and then follow their transmission micro satellites: these are short repetitive sequences that are abundantly interspersed throughout a species' genome and tend to vary in length among diff individuals Usually contain di, tri, tetra, or pent nucleotide sequences that are repeated many times in a row Ex: most common one found in humans is dinucleotide sequence (CA)n- where n ranges from 5 to more than 50!!! So repeats CA up to 50 times!! CAn microsattelite is found on avg about every 10000 bases in human genome Have identified thousands of diff DNA segments that contain Can locates at many distinct sites w in human genome— so appears in lots of our cells How do researchers identify specific micro satellite w in chromosome? 1- have set of chromosomes and add PCR primers 2- the PCR primers specifically recognize sequences on chromosome 2 that flank a particular micro satellite — so the primers flank to either side of a microsattelite Primers complementary to the unique DNA sequences that flank a specific region are used to amplify a particular micro satellite by PCR— the PCR primers copy only a particular micro satellite but not the thousands of others that are interspersed throughout the genome!!— bc the primers go around regions of specific ones! Sequence tagged site (STS)- site where a pair of primers copies a single site w in a set of chormoeomss - the amplified region is called this 3- many cycles of the PCR produce a large amount of the DNA fragment contained bw the 2 primers (WHERE THE MICROSAT IS LOCATED)— -- have forward primer and reverse primer that work towards each other until hit micro satellite region then goes through gel electrophoresis AND the microsatellites w more repeates- longer so will be more towards top and w fewer move farther down gel— so individual could be heterozygous for a microsattelite on specific chrom- here #2 and gel will allow to see how the 2 copies differ in length!!— if homo- 2 bands will be same on gel SO 1- amplify DNA through PCR to produce large amount of the fragments contained bw 2 primers-- if there is a lot od DNA contained bw the 2 primers-- a lot of repeats- then the fragment w micro satellites w more repeats will stay at top of gel **if sequence individual- both of their chromosomes and get 2 bands on gel-- hetero!!

linkage studies and following transmutation of micro satellites through generations' ex w pedigree - look at pic in notes

micro satellites as polymorphism- the polymorphism in micro satellites is based on the length of fragment !!!!! B= longer frag, more repeats b= shorter THEN - follow transmission

mito DNA- was perf for finding common ancestor- mito eve but- to conclude hypothesis on mixing of pops? what need?

mitochondrial DNA- no crossing over, recombination bc not inheriting from both mom and dad- so cant see interbreeding the common ancestor- far away- good to use mito now want to see interbreeding (more recent evo history) NEED whole genome sequencing to determine interbreeding Whole Genome Sequencing reveals interbreeding with other ancient homininds can see- old old ancient species earliest to break right into asia- Denisovans then neadrathals break (independent of Denisovans into Europe) homosapiens break their-- coming from Africa- mix w both ! Closest to modern is at top and can see humans found in every continent but if look way back there is places of time where we overlapped w other common ancestors like neandrathals- so could have been interbreeding bw modern humans and these other types of people Can see this in autosome! Denisovans: Ancient hominids discovered in Siberia neandrathals and humans- share more recent common ancestor we see some Neanderthal sequences in humans- bc were in same place at same time eventually- humans first collided w neadrathals- then w denisovan- so eventually as spread out across continent: Many human lineages have Neanderthal and/or Denisovan sequences

how confirmed gene four melanin study and understanding melanin and melanocytes and keratoniytes also location of melanin?

model system = mouse Model system here is mouse melanocytes- these mouse melanocytes are grown in cell culture **can zoom in on cell melanocyte bc know trait of interest is the melanin which is made inside ** so expression of the protein involved must be turned on in the DNA in these cells! Understanding the cells: They are really interesting shaped cells that produce melatonin in vesicle then they bud off and get abosirbed by keratinocyte and then when sun hits keratinocytes if melon arranges around nuclei of these on Side that will get hit by light- assemble them around nucleus on side that will protect cells ** THIS MAKES SENSE- more melanin in darker skin ppl bc need to cover keratinocytes more to protect from sun coming in and not degrade! FIRST: looked at actual gene of MFSD12 to see if it was involved in production of melanin Used ShRNA knock down - uses short hairpin ran - ran base pairs w self to form hairpin loop-then w in cell protein called dicer that cuts off tops of loop- now have double strand RNA— then second protein complex RISC will base pair the ran w the MFSD12 mrna that will base pair w mrana coming I — then simulating mutation by degrading the mrna- so not enough protein will be made results: when put in high amount of ShRNA- silenced mrna of MFSD12- so low amounts of MFSD12 and pigment observed= dark!! Mcan see- in normal amount of MFSD12 - normal amount of expression then when MFSD!@ is silences— more melanin The ShMFSD!@ look darker — then also could number of melanosomes w in cells and found more Looked for overlap— red +green= yellow and yellow indicates that MFSD12 is overlapping w lysosomes and not w melanin Of separate then MFSD12. Is not in melon MFSD12 doesn't overlap w locations where melanosomes are found but does overlap w lysosome marker SO conclusion is that it Is in lysosomes but NOT melaanosomes This is suprising!!— doing something in lysosomes second model system : Second model system- zebra fish- 2 versions, one ecpresseding MFSD and one not expressing Here looking at production of yellow pigments more similar to pheomelanin **wanting to understand relation of gene to both pigments (light and dark, eumelanin and pheomelanin Here looking at production of yellow pigments more similar to pheomelanin Top panels show- less pigment in not expressing MFSD Bottoms are control- shows cells are there they are just not making ass much pigment SO looks like if lose MFS12 or corollary in zebra fish, less pheomelanin Another model system: aguti mouse- use CRISPR to destroy MFSD12 gene- produces gray mouse So in mice gene product MFSD12 is involved in pigment production

Multi Gene traits and penetrance- Forward genetics ex w melanin tests on next slide!!

pigment forward genetics (jump forward to go back) observe phenotype then look back to find gene that causes it steps: 1: find a population to study that has diversity in the trait of interest 2: categorize each individual for that trait 3: perform sequencing or other methods (SNP,SSLP, RFLP) 4: statistical analysis to look at markers that are more associated with the trait than would be by chance 5: Examine the genome around markers that are associated with the trait for genes and sequences that may control the trait eumelanin: brown pigment pheomelanin: red and yellow differences in UV at diff locations : Botswana, Ethiopia, tanzania - there are distinct ethnic groups in each place so most likely expect to find multiple alleles at multiple sites 2: group individuals- did this with melanin index- amount of melanin in skin (melanin= pigments produced in special group of cells called melanocytes - melanin absorbs light) 3: find genes associated w traits -- can use sequencing or other methods to detect molecular markers here: SNP array - did a 6 million SNP array- 4.2 million were informative remember the genes are focuses - the MFSD12 is a locus- so may have some hits on gene itself or promoters by it **BASED ON RELATIVE INDIVIDUAL VALUES OF MELANIN STAT ANALYSIS ALLOW FOR THIS: SAY YOU HAVE 3 INDIV 1 AND 2 SHARE A PLACE IN GENOME AND 3 AND 4 SHARE A PLACE IN GENOME - ALL DARK SKIN BUT TO FIG OUT WHICH GENE CAUSING DARK SKIN- BW 3 - WHICH 2 DARKEST- THE SHARING OF THAT GENE PROLLY MORE INVOLVED IN SKIN COLOR graph: y axis is function of prob- higher up on y - more prob that gene is aossicaed non randomly w trait - found 4 regions of high prob 5: examine the genome around markers that are associated with the phenotype looked at MFSD12 locus- chromosome 19 This is the zoom in of the place in chromosome 19 that was indicated to have a polymorphism in length o sequence compare the SNP and prob chart- put on top of genes draw lines from high dots on SNP array (high prob of non random association) to genes and found so high p value parts align w parts of locus right bf exon and some align w part of locus on exon!! ALSO: 2 diff types of cell type that contain melanin melanocytes and keratinocytes This is all forward genetics: start w phenotype of interest - find gene (or several genes) In this case- researchers found 4 genes involved in skin color. What would be next step you would take? Complementation test- tells u whether 2 diff mutations are in same gene ( we already now that they are bc mapping to diff places in genome so we already know this!) *would perform these if wanting to see if — also cant mate too humans together! This is humans don't do this Epistasis analysis- could look at existing pedigrees w in these populations - but w in family probably have same SNPs — also would need to cross- so don't do this

When scientists study populations, they typically observe high levels of variation in many of the traits that characterize a species. In genetics, variations in traits at the population level are known as

polymorphisms -- genes are polymorphic if in pop have relatively same amount of freq Consider a population, for which you know the allele composition of three genes. Gene A has three alleles in the following proportions: A1 - 34.4%, A2 - 30%, A3 - 35.6%. Gene B has two alleles in the following proportions: B1 - 67.1%, B2 - 32.9%. Gene C has two alleles in the following proportions: C1 - 99.2%, C2 - 0.8%. Which of those genes is(are) polymorphic? A and B only This does NOT have to be true for HWE Immigration and emigration rates must be equal, resulting in no net movement of organisms.

Evolution and detecting genetic variation - lecture 25 pop genetics defintion approaches to understanding evolution of alleles Hardy weinburg eq and SO WHO CARES? ex w sickle cell

population genetics- how alleles or genes change over time at the population level - observing genotype and phenotype beyond family trees approaches to understanding evolution of alleles 1: sample DNA sequences over multiple generations 2:compare sequences from existing populations to archaeological samples or fossils 3: look for evidence of change based on deviation from equilibrium hypothesis equal hypothesis (hardy): says allele freq (NOT GENO) changes if evo is occurring! But if looking at DNA sequences- this is molecular evolution because DNA is a molecule Equal hypothesis- start w baseline hypothesis that alleles and genes will behave a certain way then if see deviations can try to figure out why deviations HWE Calculates genotype frequencies based on allele frequencies Describes how a population will change over future generations based on current allele frequencies if the conditions of equilibrium are in place - so get p and q value descibes how pop will change if current allele freq not changing (not undergoing evo) it assumes no evo so no genetic drift no sexual selection no natural selection infinitely large pop size no mutations no fitness advantage for any allele equal mating among males and females no migration *in general: its a way we can calculate predicted genotype frequencies (what is interesting to us) based on allele frequencies if no evo! Used to calculate genotype frequencies based on allele frequencies- Can described way population is or how it will change based on what we know about current state of how equilibrium is Assume no difference in fitness or selection of alleles- no one allele is more beneficial than the others— think of this in terms of SNP's- in space of genome, polymorphic site- not causing specific phenotype maybe SNPs aren't messing w anything! Given a population with two alleles at a given site, G and g: The frequency of G alleles = p The frequency of g alleles = q Prediction is that the allele frequency will not change over generations In random mating, (p + q) can describe the possible gamete contents made by the population. When two gametes come together to form a diploid, (p + q) x (p + q) = 1 all individuals of the next generation will be made from one gamete and another gamete (p + q)2 = p2 + 2pq + q2 = 1 genotype frequencies can change!! Can predict the occurrence of carriers based on the number of affected individuals CDC says 100,000 Americans have sickle cell disease HbS HbS genotype (330 million Americans) According to Hardy-Weinberg, what is the estimated frequency of carriers of the HbS allele? whats cook about it- don't have to have population where everyone is genotypes In many treats easier to observe homozygous recessive- so could observe that and then of the HOMO Dom vs hetero- harder to tell but now can know how many carriers vs homo Dom using this eq!! Can use likelihood that someone is carrier to determine whether someone in next generation will be carrier

we know something about MFSD12- now to relate back to human sequence and address other segments

sequence from the MFSD enhancer increases expression of a luciferase construct in cultured melanosomes So gene is affected but how do we take back to what is happening in human populations observed- look for enhancer activity in diff genes Some of the snips were in MFSD12 enhancer regions (amount of expression of MFSD12 impacts color we know this based on mouth studies so need to look at the) So what did was take gene and then upstream of luciferase (encodes protein that glows) put diff variations of enhancers— if glowing more then more enhancer Did this w 2 sequences associated w lighter skin See more expression of luciferase reporter gene Consistent w what thinking- if less MFSD12- more of dark brown eumelanin and what this is saying is that in cases where less eumelanin, higher expression from those variants These agree w one another- this and last studies

Using DNA sequences to study human origins and migration - models for how hominids migrated and evolved to form current populations

slocals: human like or human forms that were there bf — this is the idea that there already was an existing population of humans or human like A: An ancestor species migrated 1 mya and modern humans migrated again 100,000 years ago and replaced the locals --Model A- there was an initial migration (moving right) to diff places on planet and those diff pops existed for hundreds of thousands of years but then subsequent migration by homo sapiens replaced those migrations— Bc of subsequent modern human migrations we don't see the early migration of ancestor species (no seq remnants bc of subsequent human migration (took over) B: Same as A with some "local" genetic contributions persisting --B: again homo sapiens replaced mostly but the dotted lines show some local identity from those previous populations C: Migration among populations occurred more than twice ---C: suggest there was an early migration but if they can move out of that po they can move back- pops moving back and forth exchanging generic material among them as they cohabitated- would see mixture in lineages of sequences we see today D: Migration of an ancestor happened once. Modern populations are the result of independent evolution simpliset- one migratopobn bf homo sapiens truly form and then model pops evolved independently CONCLUSION: CONCLUSION FROM CLASS: WHICH MODEL MOST CONSISTENT? B

THEN- did gene drive- how done

w CRISPR- now when make break- guid RNAs directs cas9 to cut but then insert CRISPR construct (was 9 and guid rna) Disrupt dsx with a construct that includes cas9 and the guide RNA cas9 and guide RNA under the control of a germline-specific promoter ***BEING UNDER GERM LINE SPECIFIC PROMOTER MEANS THAT THER PROMOTION (CREATION OCCURS) ONLY GERM LINE SPECIFIC- THIS COMPLEX IS TURNED ON ONLY IN GERM CELLS --SO! MALE MAY INHERIT THE MUTATION BUT BC MALE HE IS FINE BUT WHEN HE GOES TO MAKE PROGENY - ALL HIS PROGENY GET IT SO !!- EVEN IF HETERO-- IF YOU HAVE COMPLEX YOU HAVE NO CHANCE OF PASSING ON YOUR ONE GOOD ALLELE BC IN THE FORMATION OF YOUR GAMETES THEY ESSENTIALLY TURN HOMO (THE CAS9 COMPLEX IN MUTATED VERSION CHANGES THE REG VERSION-- SO ALL ALLLES YOU PASS ON ARE MUTANT) SO IF 2 HETEROS MATE BOTH WHO HAVE COMPLEX--- ALL PROG WILL NOT BE FERTILE!! Resulting gametes made by heterozygous individuals are all dsxCRISPRh mutant