Story of Haplogroup G in India

Arvind Rangaswamy

(This version October 2019; originally posted in March 2009)




I have summarized some facts and insights about the presence of haplogroup G (specifically, with reference to the subtype G2a2b2a4a1a1 (Z30522), which is my haplotype) in India (see  Z30522 is a subgroup of G2a2b2a (P303), the largest group now known within the G haplogroup. The “story” here is based on a genographic analysis of my Y-DNA, other DNA databases, historical facts and interpretations, and personal knowledge.  There is a lot of research going on currently that will add more texture to the story in the next few years.


The G haplogroup probably originated in Eastern Anatolia (present-day Turkey, Armenia, or Western Iran), and has been present in the mountainous regions between the Near East and India during historic times (i.e., Caucasus, Northern Iran, Afghanistan, Hindu Kush), and which today constitutes only a small but widely distributed proportion of Europe’s population (about 3-4%) and the world’s population (about 1.0%), and is present in less than 1% of India’s population (see Table at the end).  The total number of men belonging to the G haplogroup in the world today is about 75-100 million.  Even 3,000 years ago, haplotypes such as G2a (a sub-group of G) and R1b (a subgroup of R that is the dominant group in Western Europe today) were widespread because many people carrying these gene markers were probably horsemen and warriors who could travel wide distances, as compared to agriculturists.  (By around 5,000 years ago, the horse had been domesticated somewhere in the Caucasus areas – riding horses in the mountains was far more practical than pulling carts with cows.  A horse could travel at least 20 miles a day, especially in the Steppes of Central Asia -- the horsemen were the jetsetters of the ancient world).  Four of the six ancient skeletons (5,000 to 7,000 years old) found in Europe so far (mainly in Germany and Spain), have now been identified as belonging to G, including the Iceman found in the Italian Alps who belonged to G2a2a1a2a1a (PF3147 branch, L166) (


Below is a map of the distribution of G haplogroup at the beginning of the current era.  The large font indicates areas where G was over 25% of the population and the small font indicates regions where G population was between 10 and 25%.




The most common G haplotype present in Europe today is G2a2b2a1 (P303).  The earlier branches, such as the G2a2b2 haplotype, are associated with the spread of agriculture in Europe, but the latter variants in Europe (and the most numerous in present-day population) are potentially a result of the spread of the Roman empire -- there is an apparent Romanian-Austrian-German-Swiss-Dutch-English axis to explain the spread of G in Europe during the period 100 – 500 CE closely following the borders of the Roman empire.  Either they were mercenaries in the Roman army, or were fighting against the Romans.  A few years ago an analysis of the presumed blood of King Louis XVI of France (yes, the one who was guillotined) was G2a2a ( and, but other genealogical studies raise some doubts about this conclusion (;, because this marker is not shared by other present-day descendants of the Bourbon kings.  Recently, it was reported that King Richard III of England had the marker G2 (, but we do not currently have information about the subtype to which he belonged.  Richard III was the last “English” king and was neither Norman nor Anglo Saxon.

Within the G Haplogroup, I have the specific markers G2a2b2a4a1a1 and I am also negative for markers L13- L140- M286- M406-. Others with related genetic structures have been found in Pakistan, Saudi Arabia, Iran, Iraq, Ukraine, and Italy.  [G2a2b2a itself is found today in high percentages among Circassians (Adyghe, Abkhaz) whose pre-historic home is believed to have been in the Northwest areas of the Caucasus (see Table at the end)].

Some observations regarding timelines for the G marker presence in India


Around 5,000 years ago, the Circassian homeland was the site of the Bronze Age Kurgan culture associated with Proto-Indo-Europeans.  It may be that the ancestor of G2a2b2a was part of the early Indo-European migrations East, arriving in North Eastern Iran/Hindukush areas during the first Millenium BCE.  But, in my assessment, G2a2b2a entered the present-day Indian gene pool only about 2,000 years ago (see below for details) and had a common ancestor with G2a2b sub-groups about 5,000 to 10,000 years ago.  I believe that my paternal ancestors have been in South India for at least around 1,000 years, and perhaps even up to 1,800 years, having some connection to Kanchipuram in Tamilnadu.  In India, the G haplogroup is present in a range of different groups, including about 64% of Afridi Pashtuns in Uttar Pradesh, 13% of Iyengars (most of whom are likely to be Z30522), 11.1% of Bihar Paswans, 11% of Gujarati Brahmins, 10% of Iyers, 6.5% of Bene Israels of Bombay, 3.3% of Maharashtra Brahmins, 3.6% of Punjabi Brahmins (as well as other Punjabis), and a few other groups with small representations: Parsis (Zoroastrian) and Ambalakarars (Kerala temple dwellers). 


Current research suggests that those living today who belong to the G2 Haplogroup had a common ancestor more than 20,000 years ago somewhere in Middle East (Lebanon, Turkey, or Iran).  The G2 haplotype (i.e., P287+ marker) appeared around 30,000 years ago.  It is likely that surviving G members are descendants of a relict population of hunter-gatherers that survived the ice age.  G2a2b2a probably emerged around 12,000 to 14,000 years ago (as per a study by Rootsi and colleagues published in May 2012, but his estimate is likely on the high side because he did not have information about subgroups of G2a2b2a (P303) that are negative for L140), and his descendants represent the dominant sub-type of the G Haplogroup today, and they are spread out across many parts of Europe and Central Asia.   Recent evidence from a Neolithic era site (5,000 BCE) in Central Europe suggests an Anatolian (Turkey, Lebanon) connection for G2a2b (L30+ or M485+) 

(;jsessionid=3E3F29E1D43A968DA524A1DE76A7043B.ambra02?uri=info%3Adoi%2F10.1371%2Fjournal.pbio.1000536&representation=PDF). Since that time, descendants of this G2a2b individual firmly established their presence in the areas South and East of the Caucasus.  The wide geographic dispersion of G2a2b2a suggests that it is an old mutation in the G Haplogroup.  Interestingly, older haplotypes, such as G* and G1 (M342+) today have limited geographical presence, primarily in Turkey, Kazakhstan and Northern Iran.  It is likely that later migrations of Scythians and Sarmatians (Northern Iranian tribes) into Europe around 2,000 to 2,500 years ago consisted of some whose ancestors belonged to G2a haplotype (note, however, that the primary haplotypes of the Scythians were likely to have been R1a1 and R1b – recent research from ancient bones suggest affiliation of people with G2a and R1b markers 


Some descendants of G2a haplotype are likely to have been present in the North Eastern Iran/Afghanistan/Hindu Kush areas for at least 5,000 years, if not longer.  They were probably nomadic horsemen, cattle herdsmen, or pastoralists, who followed their herds, or were constantly looking for locations with good rainfall for their crops.  The cow was a transportable asset, and a source of food.  Starting around 4,200 years ago, there were increasing migrations into present-day Northern India.  One potential reason for these migrations was the changing climate (see, for example, Drought in 2,200 BC), which induced substantial migrations from the Indus valley into present-day India, as well as migrations of Indo-Iranian (Aryan) tribes from the Northwest via the Hindu Kush Mountain.  Unlike the people of the Indus valley civilization, the newcomers from the Northwest did not know anything about irrigation. 


Although some with the G2a2b2a marker could have come into present-day India around 3,500-4,200 years ago along with the various early migrations of the Indo-Iranian tribes into the Gangetic plain, I think a later entry around 1,800 to 2,200 years ago is much more plausible, when there were additional migrations of Saka tribes (by this time, the Indo-Iranian tribes were called Sakas, or in today’s terminology, Indo-Scythians) from present day Afghanistan as they were pushed South by the Chinese Yuezhi tribes, and later, by Kushan kings. The Indo-Scythians migrated to parts of Punjab, Gujarat, and West-Central India, and from there later on to Southern India, Ceylon (present-day Srilanka), and also onto Cambodia. These same pressures from the Chinese tribes may have encouraged others with the G2a2b2a marker to move west, and eventually to the Caucasus region and Western Europe, where they are present today in substantial proportions among certain ethnic groups.  G2a2b2a’s entry into India during a time later than the early Indo-European migrations is supported by some recent research results.  (1) Sanghamitra Sengupta et al. 2006, and Swarkar Sharma et al. 2009 find that G2a is present only in 0.9% of the Indian population, compared to the 21% for the R1a1 haplotype, more directly associated with the earlier Indo-European migrations.  (2) G2a is found in high frequencies (10% or more) in only five reported groups in India, namely, Iyer and Iyengar Brahmins, Gujarat Brahmins, Afridi Pashtuns, and Bihari Paswans.  More interestingly, it is not present in many other groups associated with a strong presence of R1a1 haplotype, including Rajputs, UP Brahmins, Bengal Brahmins, MP Brahmins, and Konkanastha Brahmins.  (3) The total number of R1a1 in India is about 120 million, compared to 2 million G2a.  It would require 60-65 additional generations (about 1,800 to 2,100 years) to grow the 2 million G2a males into 120 million male descendants, if on average, each G2a has 2.1 children (the average growth rate in the past 2,000 years of human history).  The formula for computing the number of male descendants after n generations is given by the well-known equation: P(n) = [cn-x+1][cx – 1]/[c-1]. Here c is the average number of children divided by 2 (=1.05), x is the number of generations that is alive at any time (we can assume this to be 3), and we can assume one generation every 30 years. 






Journey of my paternal ancestor into South India


I belong to the Iyengar community in South India (worshippers of Vishnu) whose members traditionally served as priests in temples.  I believe there are three distinct possibilities for how my paternal ancestor (with the Z30522 marker) could have reached Tamilnadu.  I have listed these in order of their likelihood of occurrence (future research should provide more clues to select among these options):


1.      My ancestor may have come as a Roman trader, or as one of their representatives, likely from Egypt/Saudi Arabia via a sea route.  There was a substantial amount of regular trade between the Roman empire and South India starting 2000 years ago during the reign of Augustus:  Archeological remnants of a Roman settlement has been found in Arikamedu (Tamilnadu). This seems to be the most plausible scenario for my ancestor’s presence in South India because of three important reasons: (1) The gene marker Z30522 is a branch of G2a2b2a4 (Z30503) which is present in Italy with sub-branches in Saudi Arabia, India and Pakistan.  (2) Early Roman trade centered around luxury items such as pepper, perfumes, and precious stones.  In particular, the best quality of beryl in the world was mined in Padiyur in Coimbatore district, a district where the largest hordes of Roman coins in India have been found (*.html).  Interestingly, I do know that my great-great-grand father was a native of Kangayam, which is near Padiyur. (3) When I examine my “distant cousins” listed at 23andme (over 400 relatives) most were either South Indians (mostly Iyengars or Iyers) or were of European origin.  I saw few if any names from Gujarat, Punjab, Pakistan, Iran, Iraq, etc. listed as distant relatives (see below for other possible reasons for the presence of the G haplogroup in South India, which involve geographic transitions through present-day Pakistan, Punjab, or Gujarat).  In particular, unlike my gene marker, Gujaratis within the G haplogroup typically have the G1a1b marker, Punjabis have G2a1a2 or G2a2a1a2a1a markers, and Pakistanis have G2a1a2, G2a2a1a2a1a, G2a2a1b, and G2a2b1a2. 



An example imitation Roman coin minted in South India (from my collection). These coins were in circulation around 4th Century CE (the head of a laureate on one side, and cross on the other side):

Ancient Coins - INDIA, ROMAN: Unattributed copper coin with cross. Rare.


2.      It is also possible that my ancestor came via a land route from present day Iran/Afghanistan. He  could have been part of the Indo-Scythian (Saka) empire or the related Indo-Parthian empire, perhaps serving as a priest in temples built by their kings, or serving in their administrative positions.  These groups have connections to the Caucasus region where haplogroup G has a strong presence.  Their journey into South India is part of the history of ancient India.  The Scythians lost their battles with Alexander the Great around 327 BCE, and subsequently were subject to further pressures from the Yeuzhi tribes starting around 150 BCE.  With the gradual disintegration of the remnants of Alexander’s empire, many of these tribes (Sakas, Indo-Parthians, Indo-Greek, and affiliated tribes, such as Kambojas (royal Scythians)) moved South toward India, and established kingdoms.  The first Indo-Scythian king was Maues (likely from the Kamboja tribe) who ruled from 85 to 60 BCE, with his capital at Sirkap, near present-day Taxila.  Gondophares I (20-10 BCE) was the founder and first king of the Indo-Parthian kingdom, with his capital at Taxila.  Gradually, the Indo-Scythian and Indo-Parthian kings extended their empire and established their capitals at Mathura for Northern satrap and in Gujarat for the Western satrap.  These kings were known to be tolerant of different religions, and adopted Indian culture, including the post-Vedic social structure based on the caste system.  The Saka kings and nobility were absorbed into the “Kshatriya” caste of Hindus.


With the decline of the Indo-Scythian empire (the last Indo-Scythian king of the Western satrap was Rudrasimha III whose reign ended in 395 CE), the Sakas and their affiliated groups migrated to several parts of South India, and also to Sri Lanka and Cambodia (presumably derived from “Kamboja”).  One possibility is that the remnants of the Western satraps of the Saka/Parthian empire intermarried with members of the Satavahana empire and migrated toward present-day Andhra Pradesh, Maharashtra and Karnataka, eventually reaching Tamilnadu and establishing the Pallava kingdom with its capital at Kanchipuram (around the end of 3rd century CE).  See, for example,  With regard to the Northern satrap, the remnants of this group are likely to be part of today’s Punjabis, with surnames Kamboj and Kamboh, and the Paswan community in Bihar (which consists of more than 40% R1a1 haplotype, in addition to 11% G2a).  The name Paswan comes from the Persian word Paasban meaning “guard.”

According to Dr. Jouveau Dubreuil, Savisakha, a Pahlava minister of Rudradaman-I (a Saka king), was the ancestor of the Pallavas of Kanchi.  According to the Velurpalaiyam copper plates (from ninth century CE), which establishes the origins of Pallava power, the Pallavas attained royal status when a Pallava named Virakurcha married a princess of a local “Naga” kingdom in present-day Karnataka or Andhra regions.  According to P.T. Srinivasa Iyengar (History of the Tamils), the first Pallava king who established himself in Kanchipuram was Bappa-Deva (this could be a generic name for “father-god,” referring to the father of SkandaVarman who is considered to be the conqueror of Kanchipuram around 280AD – see analysis by Heras in his Studies in Pallava History).  Bappa-Deva is said to have made donations to 20 brahmin families belonging to the Atreya, Harita, Bharadvaja, Kaushika, Kasyapa, and Vatsya gotras (a gotram is the paternal lineage assigned at birth).  Interestingly, this list does not include the gotram to which I belong (SriVatsa gotram), suggesting that this gotram came into being at a later date. The Pallava kings claim to be “Brahma-Kshyatriyas” (i.e., brahmins who became kshyatriyas in order to rule) belonging to the Bharadvaja gotram (Another possibility is that Pallavas were kshyatriyas whose spiritual guides were brahmins from the Bharadvaja gotram).  The early Saka kings promoted Buddhism.  Eventually, some Saka kings and their descendants started worshipping Vishnu.  During the Saka period, Buddha was transformed from a wise teacher to an incarnation (avatar) of Vishnu – the Saka king Rudradaman was known to be a worshipper of Buddha as a form of Vishnu, and married a Hindu woman.  Interestingly, the “Srivatsa” mark began appearing on the chest of Vishnu starting around this period, perhaps to indicate Buddha as an incarnation of Vishnu (Gopinath Rao T. A., Elements of Hindu Iconography, Vol. I, 1985, p. 25).



3.      A third and, as per my assessment, the least likely possibility is that my ancestors stayed in the Gujarat area following the decline of the Saka and Indo-Parthian empires, perhaps as priests serving Somnath temple, dedicated to god Shiva.  Some Zoroastrian priests (magas) of the Sakas converted to brahminism after coming to India (there is documented evidence of conversion of 18 priests).   They could have become the bhojaka priests. These converts became scholars of Sanskrit – the Indo-Scythian-rulers were the first in India to introduce Sanskrit as the official language for State-related communication (before second century CE, there is no evidence of use of Sanskrit in India for official State business).  When Somnath temple was first destroyed in 725 CE by the Arab governor of Sind, or when the re-built temple was destroyed by Mohammed Ghazni in 1024, many of the priests fled Gujarat, some to Tamilnadu, at the invitation of the Chola kings.  There is a historical record of migration of priests from Somnath temple to Tamilnadu during this period.  I believe my ancestor is not very likely to have come South via this migration.  A key reason is that the proportion of G2a2b2a among Iyers (worshippers of Shiva) and Iyengars (worshippers of Vishnu) is roughly the same (see below in the graph).   Had my ancestor come after the raids on Somnath temple, there would really be no reason for him to convert to Vaishnavism from Shaivism after migrating South, when Shaivism was already flourishing in the South.  However, my ancestor could have come with the earlier migrations of the remnants of the Indo-Scythian empire possibly as a follower of Buddha.  Some of these migrants converted later to Shaivism (as evidenced by substantial presence of G2a2b2a among Iyers, worshippers of Shiva), but some would have converted to Vaishnavism (some Pallava kings were worshippers of Shiva, and others were worshippers of Vishnu, but the Pallava kings were generally tolerant of both religions).

Based on the above considerations, here is a summary of the three possible ways my ancestor ended up in Tamilnadu, India: (1) Rome --> South India (possibly via Middle East, (2) Caucasus-North Eastern Iran-Bactria (modern Balkh in Afghanistan) 1,700 - 600 BCE -->  Ancient Kamboj/Hindukush  (600 – 200 BCE) --> Gandhara/Taxila (Takṣaśilā) (200BCE – 50CE, likely via Bolan pass) --> Gujarat  (50CE – 200CE) --> Maharashtra/Andhra Pradesh (200 CE – 300CE) --> Kanchipuram, Tamilnadu (300-800CE).  Or, (3) Caucasus-North Eastern Iran-Bactria 1,700 – 600 BCE -->  Ancient Kamboj/Hindukush (700 – 200 BCE) --> Gandhara/Taxila (Takṣaśilā) (200BCE – 50CE, likely via Bolan pass) --> Gujarat (50CE to 800CE) -->Kanchipuram, Tamilnadu (800-1000CE).


In the past 20 years, large hordes of Pallava coins have been found in Karur and Tirukoilur in Tamilnadu, India (The Pallava Coins, R. Krishnamurthy).  Some Pallava coins have also been found in Sri Lanka and Thailand.  Interestingly, numismatic records indicate the use of common Buddhist symbols such as “Swastika” and “Srivatsa” marks on coins of Parata Rajas (the Indo-Parthian/Greek governors of Gujarat and Saurashtra were called “Parata Rajas.”), the coins of subsequent kings of vassal states under the suzerainty of Satavahana kings of Andhra, and on many of the coins of the Pallava kings.  The Indo-Scythian coins, with one exception, do not have these marks, which is consistent with the notion that Pallavas are more likely to be Indo-Parthians (the following examples are from my collection). 



 Parataraja Koziya (1st century BCE)                     Chutus Karnataka/Andhra (160-345CE)                   Early Pallava coin (circa 500 CE)


Speculations about the identities of my distant paternal ancestors


I know of one (presumed) direct paternal ancestor, Nadadur (or Nadathur) Ammal (also known as Vaathsya Varadachariar of SriVatsa gotram), who lived in the 12th century (born 1165 CE) near Kanchipuram, Tamilnadu, India.  Nadadur Ammal was the grandson of the nephew of Sri Ramanuja Acharya, exponent of the Vishistadvaita philosophy of Hinduism (Sri Vaishnavism), and was himself, a leading proponent of this philosophy.  Nadadur Ammal’s paternal grandfather, Nadadur Alwan (also called Varada Vishnu Acharya) was the son of Sri Ramanuja Acharya’s sister, Kamalambal (aka Kamala Devi).  Nadadur Ammal is credited with writing 15 books in Sanskrit, including PrapannaPaarijatam (, which are available even now in a few libraries, some translated into English and German. (Sri Ramanuja Acharya belongs to the Harita gotram, had two younger sisters, and did not have children).


There were about 50 million men living 2,500 years ago (  Today, we have about 3 billion men.  On average, therefore, each male living 2,500 years ago will have 60 descendants today. However, the founder of G2a2b2a has over 2 million living descendants!  It is the largest G group in Europe, and consists of at least 1-2 million men there.  Its presence among just the Iyer and Iyengar castes in India provides another 150,000 (total Iyers and Iyengars 3,000,000; half of them are males, and about 11% of them are G2a2b2a1, or L140+).  It is not unrealistic, therefore, to estimate that there are at least 2 million G2a2b2a men in the world living today.  What this means is that the founder of G2a2b2a belonged to a very privileged group and had a large number of children (especially in the formative years of this branch of the Y-DNA).  There are only a few possible founders of G2a2b2a 2,500 years ago who could account for such observed “elite dominance.”  If my ancestor had come via Iran, Pakistan, Gujarat, then a potential founder of this haplotype could be emperors such as Cyrus, Darius, or Cambyses (Kamboujyeh in Persian), or the later Mithradates, or a religious leader like Zoroaster, or one of their immediate ancestors. 


Distribution of G Haplogroup in India and among Iyers and Iyengars of South India.


Among Iyers and Iyengars, the major Haplotypes are R1a1, J2a, J2b*, G2a, R2, H1, L1, and C5. 


The Iyengar name is an extension of Iyer, and has an interesting origin.  When the north Indian Brahmins settled in the South (mostly after 4th century CE, although there were small migrations earlier), the word “Ayya” (from “Arya”) was used by the local Tamils to refer to, or address, the Brahmins.  Eventually, it became Ayyar (a respectful form).  The name Iyengar seems to have originated in the 15th century when there was a large presence of Sri Vaishnavites in Andhra Pradesh under the patronage of the Vijayanagar kings.  The title seems to have become popular around 1490 CE after Kandadai Ramanuja Ayyan came to be called Ayyangar by the Vijayanagar king Sajuva Narasimha (the extension “Garu” is a respectful form in Andhra Pradesh).  In terms of genetic composition, there should be very little difference between Iyers and Iyengars, as can be seen in the graphs below.



It is likely that the J2* sub-group among Iyers and Iyengars are mainly from those who had inhabited the Indus Valley.  Part of the R1A1* consists of the brahmins who migrated over several centuries from the North (e.g., Uttar Pradesh) to serve as priests in temples in Tamilnadu under the Pallavas and Cholas.  The G2a’s and some portion of the R1A1* are likely to be descendants of the Pallavas and their affiliated tribes.  The others are mainly “locals” who became Iyers and Iyengars via conversion to brahminism, or immigrants and soldiers from other regions of India who were absorbed into the brahminical fold.


Recent research suggests that:


·       The R haplogroup emerged about 35,000 years ago, and R1a emerged about 15,000 years ago.  There is considerable controversy regarding the origin and spread of this haplotype, and its role in the spread of Indo-European languages.  From a genetic perspective, the “into India” theory (e.g., Klyosov, 2008) suggests that R1a1 emerged about 11,500 years ago in the Balkans, and spread to Russia, Europe, Middle East, and India, and spread throughout parts of Europe and Asia, starting about 5,000 to 6,000 years ago.  The “inside India” theory (e.g., Sharma et al. 2009) suggests that R1a1 originated in India 10,000 to 12,000 years ago, if not earlier -- today about 125 million Indian men have the R1a1 haplotype.  It spread from India/Iran into parts of Eastern Europe, Russia, and further West starting around 5,000 years ago.  Underhill et al. (2009) ( also rule out a flow from Eastern Europe to India within the past 5,000 to 8,000 years. A recent study of ancient bones conducted at the Harvard lab of David Reich favors the into India theory of Indo-Europeans starting around 3,500 to 4,000 years ago (

·       R2 is present primarily in India and Pakistan, although it originated in South Central Asia about 25,000 years ago. 

·       H1 and L1 have been in India/Pakistan for about 30,000 years.

·       R1a1, R2, H1, and L1 are widely distributed across many castes, tribes, and religions (Hindu, Muslim, and Christian) of India. 

·       C5 probably originated in India though not widely distributed in India (mainly in the South), but is also present in Polynesia and in Australia. 

·       G2a probably originated in the SouthEast Caucasus/Northern Iran area, and may have been in present in the Hindu Kush area for as long as 10,000 years.  The most likely scenario is that this haplotype entered India’s gene pool around 2,000 years ago.  

·       J2a and J2b* are of Middle Eastern/Iranian origin, which have broader distribution in India than C5 and G2a, and are present primarily in the upper and middle castes.  The spread of the J2* haplotypes is associated with the adoption of new agricultural methods during the Neolithic period 10,000 to 12,000 years ago, starting from Anatolia (present day Turkey) and nearby areas.  These haplotypes are likely to be closely associated with the Indus Valley civilization, and probably entered into the Indian gene pool 4,000 to 8,000 years ago well south of the Hindu Kush mountains.


In the years ahead, we are likely to get a more refined picture of how G evolved into G2 and to G2a and further subclades of G2a, and the migratory patterns associated with these genetic markers.  Likewise, there will be discoveries of sub-clades of R1a1, which should provide deeper insights regarding the spread of the Indo-European languages and culture, which to-date has taxed the best linguists, historians, and archeologists, apart from aggravating politicians and bigots.  Considering everything, I believe there could have been multiple migrations of R1a1, both into present-day India, and out of India at different times, which could account for its deep and widespread presence in a number of countries and regions, such as India, Ukraine, Caucasus, Poland, and Russia. Unlike language, religion, and culture which are subject to assimilation, Y-DNA is only subject to slow mutations, disappearance, or elimination.  The genes that survive and their variations across groups leave clear imprints that tell a story that is hard to discern from traditional research methods. 


The genographic data that has been compiled in the past few years by many researchers have already made the following clear:


·       There is really no scientific basis for equating genetics and race.  That is, if you are provided the genographic profile of a random Indian, you would not be able to say with any certainty to which caste or tribe that person belongs.  Conversely, if you know the race of a person, you would not be able to say what genetic lineage that person will have, although there are some exceptions.  For example, a Kashmiri Brahmin has a 75% chance of belonging to the R1a1 haplotype.  Race is a social phenomenon.  Genetics is a biological phenomenon. 

·       The Indian patrilineal pool is very diverse and cuts across language, caste, tribe, and religion.  There are more than 24 different haplogroups present in India (C, D, F*, G, H*, H1, H2, J2*, K*, K2, L, L1, O*, O2a, O2a1, O3, O3e, P*, R*, R1, R1a, R1a1, R1b3, R2).  On the other hand, the Indian mitochondrial DNA pool (female ancestry) falls pretty much into just three types (M, R, U), attesting to how closely related all Indians are to each other (

·       Humans share 99.9% of their genes with each other (only 3 million base pairs differ between two randomly selected individuals out of a possible 3 billion base pairs), and humans share around 94% of their genes with chimps.  Thus, most of the genetic differences between people are superficial.  However, genographic profiles provide a way for us to understand our own origins and the migratory path of our ancestors (they may also be useful for understanding potential susceptibilities to certain diseases among people with different genetic lineages).


Worldwide distribution of G Haplogroup


Below is a summary of the distribution of the G Haplogroup across the world today.  Further details about the geographic distribution of G is available at


Ethnic Group or Country

% G

Madyars (Kazakhstan) -- G1/M285


Shapsugs (Caucasus)  G2a2b2a


Ossetians -- G2a1a1a1a1a (FGC750)


Afridi Pashtuns in UP, India M283


Abkhazians (Georgia)


Terek Cossacks


Rutulia (SW Russia)


Lezinians (SW Russia)






Kabardinian (G2a)




Abkhazians (Georgia) – G2a2b2a


Kalash (Pakistan) – (G2a)




Ibiza (near Spain) – G2a2b2a


Bakhtiaris (Iran)


Sardinia (Italy)


Gaugauz (Moldova)




India (Iyengars) -- G2a2b2a4a1a1 (Z30522)


Druze – G2a2a1a2 or L91


Mazandarani (Iran) – G2a2b2a


India (Bihar Paswans)


Assyrians in Iran (G2a2b2a)


Iran (all G)


Pathan (Pakistan) – M277/G2b1




India (Gujarati Brahmins) – G1a1b




Italy (G2a)


India (Iyers) -- G2a2b2a4a1a1 (Z30522)


Israel (Jewish population)


Gilakis (Iran) – G2a2b2a


Lurs (Iran) – G2a2b2a


Turkey (G2a)




Portugal (G2a)




Ashkenazi Jews – Various G groups






Austria (G2a)


Bene Israel Jews of Bombay




Greece (G2a)


Chechnya and Ingushetia








Czech Republic (G2a)


Spain Non-Basque






Switzerland (G2a)


Uygur (Central Asia)




Wales (G2a)


Iran – (G2a2b2a)


India (Punjab Brahmins)


Pakistan (Overall)


Ukraine (G2a)




Hungary (G2a)


India (Mahratta Brahmins)


Afghanistan Hazaras (G2a2b2a)


Saudi Arabia (mainly G2, but some G1)




Berbers (N. Africa)




Netherlands (G2a)




Ashkenazi Jews – (G2a)




India (Kashmir Pandits)




Belgium (G2a)




Spain -- Basque (G2a)


India (Overall G)



(1) The table above was compiled from several research publications including sites maintained by Ray Banks ( and  Special thanks to Ray Banks, Rolf Langland, and Ted Kandell for their insights.  Some of these percentages have been computed from non-representative samples (especially for country estimates) and are subject to revision with new data.

(2) The data for Afghanistan is from a paper by Zalloua and his colleagues (  In their data, G2a2b2a (P303) is present only among Hazaras, and not present in other ethnic groups in Afghanistan, including Arab, Baluch, Norestani, Uzbek, Pashtun, and Tajik.  The Hazaras of today live in an area where Zoroastrianism and Buddhism once flourished. It is likely that Hazaras of today were originally composed of many distinct ethnic groups, rather than being a single group with gene admixture.

(3) The data for Iran is from a paper by Grugni and his colleagues ( 

(4) An interesting observation in the Table is the presence of G2a2b2a in significant proportion in the island of Ibiza (near Spain), where migrations have occurred only in historic times, perhaps via Sephardic Jewish merchants.

(5) G2a2b2a (P303) is typically not the dominant haplogroup among various ethnic groups such as Gilakis, Mazandarani, Lars, Hazaras, Iyengars (except that it is one of several larger haplogroups within the Mazandaranis).  In these groups, other markers such as R1b1, J2a, R1a1, and J1, J2a are present in far greater proportion than G.  Further, the dominant genetic markers vary across these ethnic groups.  One interpretation of such a pattern is that those with the G2a2b2a marker were external invaders, who then assimilated with these local ethnic communities many centuries ago.  Alternatively, G2a2b2a could have belonged to the original group, and later groups mixed in and replaced the original group members.