Gene Conservation Laboratory

Publications - Abstracts

Sockeye Salmon

Seeb, L. W., Habicht C., Templin, W. D., Tarbox, K. E., Davis, R. Z., Brannian, L. K., and Seeb, J. E..Genetic diversity of sockeye salmon (Oncorhynchus nerka) of Cook Inlet, Alaska, and its application to restoration of populations affected by the Exxon Valdez Oil Spill. (Accepted pending revisions Transactions of the American Fisheries Society.)


Genetic data from sockeye salmon (Oncorhynchus nerka) were collected from the Kenai River, a major salmon-producing system affected by the Exxon Valdez Oil Spill, as well as all other significant spawning populations that contribute to mixed stock harvests in Cook Inlet, Alaska. The products of 29 enzymes encoded by 67 protein loci were resolved from samples from 47 spawning locations in Upper Cook Inlet. Allozyme data revealed a substantial amount of genetic diversity among populations. Mixed stock analyses using maximum likelihood methods with data from 27 loci were evaluated to estimate the proportion of Kenai River populations in Cook Inlet fisheries. Simulations indicate that Kenai River populations can be identified in mixtures at a level of precision and accuracy useful for fishery restoration and management. Samples from fisheries were analyzed both inseason (within 48 h) and postseason. The contribution of Kenai River populations to the Cook Inlet fisheries varied from 16.4% to 90.9%. Samples from fish wheels on the Kenai, Kasilof, Yentna, and Susitna rivers were analyzed to check the adequacy of the baseline. Results from this study are currently being used in the management of Cook Inlet sockeye salmon populations affected by the Exxon Valdez oil spill.

Chum Salmon

Seeb, L. W. and P. A. Crane. 1999. High genetic heterogeneity in chum salmon in Western Alaska, the contact zone between Asia and North America. Transactions American Fisheries Society. 128: 58-87.

Seeb, L. W. and P. A. Crane. 1999. Allozymes and mitochondrial DNA discriminate Asian and North American populations of chum salmon in mixed-stock fisheries along the south coast of the Alaska Peninsula. Transactions American Fisheries Society. 128: 88-103.

Seeb L.W., P. A. Crane, E. M. Debevec. 1998. Genetic analysis of chum salmon harvested in the South Unimak and Shumagin Islands June Fisheries, 1993-1996. Regional Information Report No. 5J97-17, Alaska Department of Fish and Game, Anchorage, AK.

Executive Summary

Genetic stock identification was used to estimate the origin of chum salmon Oncorhynchus keta harvested in the South Unimak and Shumagin Islands fisheries in June.

Over 14,000 chum salmon were sampled from fish caught during the June commercial fishery at South Unimak Island from 1993-1996 and the Shumagin Islands from 1994-1996.

The 1993 South Unimak fishery was stratified into 2 periods, June 13-20 and June 22-29 and the South Unimak and Shumagin Islands 1994-1996 fisheries were stratified into three periods, the opening of the fishery - June 20 (Period 1), June 21-25 (Period 2), and June 26-30 (Period 3). For each fishery, 400 fish were randomly subsampled from the total number of fish collected in each period, proportional to the daily catch. Genetic variation was assayed at 20 allozyme loci on the subsampled fish and used to estimate stock composition for the South Unimak and Shumagin Islands fisheries.

The allozyme baseline compiled by Seeb et al. (1995) was expanded with allele frequency data for 83 populations from Western Alaska, Canada, China, and Russia. The updated baseline comprised 248 collections of chum salmon that were condensed into 109 pooled population groupings. A multidimensional scaling analysis and simulation study on the pooled population groupings suggested 10 reporting regions could be identified in mixtures: 1) Japan; 2) China/Southern Russia; 3) Northern Russia; 4) Northwest Alaska summer; 5) Fall Yukon; 6) Alaska Peninsula/Kodiak; 7) Susitna River; 8) Prince William Sound; 9) Southeast Alaska/Northern British Columbia; and 10) Southern British Columbia/Washington (Figure 1).

Stock contributions for each period of the June South Unimak fishery samples, 1993-1996, and June Shumagin Islands fishery samples, 1994-1996, were estimated via maximum likelihood. Ninety percent confidence intervals were computed for all regional estimates from 500 bootstrap resamples of the baseline and mixture. Annual contribution estimates were calculated as a weighted average for each year with the contribution estimates for a period weighted by the total catch for that period. Ninety percent confidence intervals were also calculated for the annual contribution estimates.

Annual contribution estimates indicated that Northwest Alaska Summer was the largest individual contributor to both the South Unimak and Shumagin Islands June fisheries. Annual contributions ranged from 0.40 to 0.65 in the South Unimak fishery, from 0.36 to 0.52 in the Shumagin Islands fishery, and from 0.38 to 0.60 in the two fisheries combined. In two of the three years where both fisheries were sampled, the South Unimak fishery had a larger proportion of Northwest Alaska Summer fish than the Shumagin Islands fishery. Asia was the second largest contributor to these fisheries, followed by Alaska Peninsula/Kodiak. Estimates for China/Southern Russia, Fall Yukon, Prince William Sound, and Susitna River indicated these reporting groups were a small component or were absent in the fisheries sampled.

Variation in reporting region contribution was evident within years in each fishery. Reporting regions that made especially strong or small contributions in a year did so in both South Unimak and Shumagin Islands fisheries.

The Japanese component of the Period 2 stratum of the South Unimak and Shumagin Islands fisheries, 1994-1996, was verified with an independent genetic marker. Period 2 fish were assayed for variation at the ND5/ND6 region of mitochondrial DNA (mtDNA); the relative contribution of JAPAN was estimated using maximum likelihood and the mtDNA baseline of Park et al. (1993). Estimates based on allozymes and mtDNA for Japan were concordant and, in five out of six cases, statistically indistinguishable.

Chinook Salmon

Crane, P. A., W. D. Templin, L. W. Seeb. 1996. Genetic stock identification of Alaska chinook salmon. Final report of the Alaska Department of Fish and Game pursuant to National Oceanic and Atmospheric Administration Awards No. NA46FD0356. Regional Information Report No. 5J96-17, Alaska Department of Fish and Game, Juneau.

Executive Summary

Identification of the origins of chinook salmon captured as bycatch in fisheries targeting groundfish in the Gulf of Alaska and Bering Sea/Aleutian Islands is a management and conservation concern. Mixed-stock analysis using genetic data has been successfully used to identify stock components of chinook salmon mixtures in Washington and British Columbia and may be an ideal tool for identifying stock of origin of bycaught chinook salmon in Alaskan waters. Though populations of chinook salmon from California to British Columbia have been genetically characterized, data describing Alaskan populations are limited. In this study we collected genetic data from wild-spawning and hatchery populations of chinook salmon from throughout Alaska to better identify populations that may be contributing to bycatch in the Gulf of Alaska and the Bering Sea. We also developed a multiplex screen to assay genetic variation at microsatellite loci, a class of DNA markers. With the allozyme data, we performed simulation studies using maximum likelihood methods to test identifiability of regional stock groupings of chinook salmon in mixtures. Data were included from throughout the North American range of chinook salmon. Eight regions were studied: 1) Western Alaska; 2) Southeast Alaska; 3)British Columbia: non-Fraser River; 4) British Columbia: Fraser River; 5) Puget Sound; 6) Washington Coastal; 7) Columbia River; and 8) California-Oregon. The results of the simulations indicate that major regional groups of chinook salmon can be identified in mixtures with a high degree of accuracy and precision.

Pink Salmon

Seeb, J. E., C. Habicht, W. D. Templin, L. W. Seeb, J. B. Shaklee, F. M. Utter. 1999. Allozyme and mtDNA variation describe ecologically important genetic structure of even-year pink salmon inhabiting Prince William Sound,Alaska. Ecology of Freshwater Fish 8:122-140.


Allozyme and mitochondrial DNA (mtDNA) data were obtained from pink salmon throughout Prince William Sound, Alaska, from two hatchery, five upstream, and 20 tidal locations distributed among five management regions collected during 1994. Screening for allozymes included 66 loci for 92 to 100 fish per sample. Thirty-four loci had variant allele frequencies >0.01 in one or more collections and were used for population analyses. Eight haplotypes were detected after screening 40 fish per collection for variation at the ND5/ND6 region of mtDNA using six restriction enzymes. Significant and apparently stable differences detected by both data sets permit rejecting a null hypothesis of panmixia and support managing native populations in Prince William Sound at the regional level. Distinctions between upstream and tidal collections were detected within Lagoon Creek (allozymes) and Koppen Creek (mtDNA). Significant regional heterogeneity was detected within upstream (allozymes and mtDNA) and tidal (allozymes) collections; however, upstream collections were more divergent from each other than were tidal collections. The absence of distinction of Armin F. Koernig Hatchery from almost all regions was consistent with multiple origins of this stock. Conversely, Solomon Gulch Hatchery in the East Region was distinct from all regions but East, consistent with a more restricted origin and influence.

Olsen, J.B., Seeb, L.W., and Seeb, J.E. Genetic variation at microsatellite loci in North American odd-year pink salmon. Submitted to Transactions of the American Fisheries Society.


We examined genetic variation at five microsatellite loci in 12 odd-year populations and one even-year population of North American pink salmon (Oncorhynchus gorbuscha). The degree of polymorphism varied widely among loci. The total number of alleles for all odd-year populations varied from 4 (Oneµ3) to 58 (Ssa85), and the mean observed heterozygosity ranged from 0.42 (Oneµ3) to 0.88 (Ssa85). Microsatellite Ssa197 exhibited significant heterozygote deficiency in 8 of 13 populations and incomplete transmission of parental alleles, suggesting the presence of a null allele(s). Chi-square analysis using microsatellites Oneµ3, Ots1, µSat6O and Ssa85 revealed significant heterogeneity in allele frequency among all pink salmon populations, between odd- and even-year lineages, between all odd-year pink salmon and among pooled populations from six geographic regions. The fixation index, (q = 0.022), indicated weak but significant (p<0.01) support for sub-population structure within odd-year pink salmon. Three hypothetical pink salmon phylogenies were derived using different models of gene evolution to infer the genetic relationships among populations. The two phylogenies based on genetic drift and drift/mutation models suggest the relationship among odd-year populations is organized latitudinally. The third phylogeny, based on the step-wise mutation model, showed no north to south orientation of odd-year populations, but did exhibit strong differentiation among odd- and even-year lineages.

Rainbow Trout

Allendorf, F. W., J. E. Seeb, K. L. Knudsen, G. H. Thorgaard, and R. F. Leary 1986. Gene-centromere mapping of 25 loci in rainbow trout. J. Hered. 77:307-312.


We have mapped 25 enzyme loci in relation to their centromeres by half-tetrad analysis in gynogenetic progeny from three hatchery strains of rainbow trout (Salmo gairdneri). The observation of close to 100 percent heterozygous gynogenetic diploids for five unlinked loci suggests that near-complete interference is common in rainbow trout chromosomes. Estimates of gene-centromere recombination rates in triploids are similar to the estimates in gynogenetic diploids. Thus, the near absence of homozygotes at these loci cannot be explained by reduced survival of homozygotes. Furthermore, we found no evidence in 69 comparisons of departure from equality of the two homozygous classes, as would be expected if an allele was linked to a recessive allele that reduced survival. We also found surprisingly little evidence of differences among families within strains or among strains. The average proportion of heterozygous progeny for all loci is 0.56; this corresponds to a fixation index of 0.44 after only one generation. However, the high frequency of loci at which almost all gynogenetic progeny are heterozygous indicates that it is infeasible to produce homozygous lines by gynogenesis involving retention of the second polar body in rainbow trout.

(Abstract reprinted with permission from Oxford University Press, Journal of Heredity. The journal home page can be accessed at

Seeb, J.E, C. Habicht, and G.D. Miller. 1993. Use of triploids for gene conservation of salmonids. In: M.R. Collie and J.P. McVey (Editors) Interactions Between Cultured Species and Naturally Occurring Species in the Environment. Proceedings of the twenty-second U.S.-Japan Aquaculture Panel Symposium. Alaska Sea Grant Report AK-SG-95-03.


The use of triploid fish has been promoted for many management applications. One advantage of using triploid salmonids for stocking projects in long-term conservation of biodiversity. Stocking diploids can result in unwanted hybridization, predation, or competition with native species. Triploids are reproductively sterile thereby eliminating the potential for hybridization. In many salmonids, sterility also means that fish will live longer, resulting in some large, trophy individuals. If competition or predation problems arise between stocked triploids and native species, then halting stocking would eliminate these interactions within one life cycle.

Interestingly, proposed uses in aquaculture include gene conservation as well as enhanced meat production. Nations have considered requiring that farmed salmonids be sterilized in order to protect wild stocks from introgression due to large-scale escapes. Sterilization through induced triploidy has also been suggested as a prerequisite to the commercial use of transgenic fish proposed by some workers (e.g., see Smitherman and Dunham, this symposium).

Problems with triploid production and performance occasionally have been reported, but these problems are not consistent and depend on the species examined and on the production environment. For example, achieving 100% triploids in production lots can be an elusive goal with some species, hampering conservation applications. Additionally, we have observed triploids exhibiting growth superior to, equal to, or inferior to the corresponding diploid controls depending upon species tested and rearing environment. Finally, some workers have expressed concern that triploid males may try to spawn with diploid females in wild-stock areas, disrupting wild-stock production.

These factors make it difficult to form generalizations concerning usage of triploids. Successful applications will depend upon project goals, and managers must balance the needs of gene conservation against possibility of reduced performance of triploids (in some cases). Thought must be given to standardizing certification of triploid salmonids for stocking, and the consequences of false spawning of triploid males must be considered.

Finally, managers must be aware that the risks of stocking diploids are seldom evaluated as thoroughly as are the risks of stocking triploids. Clearly, in some cases, use of the former will have more dire consequences than the use of the latter. We believe that, given the information available now, use of triploids should be required for some types of stocking programs and most net-pen farming.


Merkouris, S. E., L. W. Seeb, and M. C. Murphy. 1998. Low levels of genetic diversity in highly exploited populations of Alaskan Tanner crabs, Chionoecetes bairdi, and Alaskan and Atlantic snow crabs, C. opilio. Fishery Bulletin 96:525-537.


We used allozyme analyses to investigate genetic variation among commercially exploited populations of Chionoecetes bairdi (Tanner) and C. opilio (snow) crabs in Alaskan waters. Data were collected from 34 presumptive loci in 1002 C. bairdi and 539 C. opilio sampled throughout the commercially important range of each species in Alaska. Average observed heterozygosities were 0.027 for C. bairdi and 0.013 for C. opilio. Low levels of geographic differentiation were detected among populations of C. bairdi and C. opilio, and our data suggest that subpopulations of C. bairdi exist within the Bering Sea. Further, evidence of gene introgression was found between C. bairdi and C. opilio in the Bering Sea. We also included a geographic isolate, North Atlantic C. opilio, in the analyses. Little differentiation was found, and no private alleles were detected in North Atlantic C. opilio despite significant geographic separation from Alaskan C. opilio.

Seeb, J. E., G. H. Kruse, L. W. Seeb, and R. G. Weck. Genetic structure of red king crab stocks in Alaska facilitates enforcement of fishing regulations. Pp. 491-502 Proceedings of the International Symposium on King and Tanner Crabs, Anchorage, Alaska, USA, November 28-30, 1989. Alaska Sea Grant College Program, Fairbanks.


Horizontal starch-gel electrophoresis of proteins has proven to be a powerful tool for the management of many marine species. This technique provides data on the genetic relationships of reproductively isolated stocks, thereby helping scientists to optimally manage these self-recruiting stocks. Additionally, when large genetic differences are found between stocks, collections from unknown origin may be genetically screened and unambiguously classified.

We examined collections of red king crab from thirteen localities in Southeast Alaska, the Aleutian Islands, and the eastern Bering Sea for genetic variation at 42 protein coding loci. Two highly polymorphic loci, Pgdh (Phosphogluconate dehydrogenase) and Alp (Alkaline phosphatase), were useful for discriminating stock differences between major geographic areas. The eastern Bering Sea collections from Bristol Bay and Norton Sound were very different from all other collections. Further, southeast Alaska collections appear to form a stock unit discrete from the Kenai, Alaska Peninsula, and Aleutian collections. Additional polymorphic loci appear to be useful in further differentiating stocks, and we are continuing our study.

In January, 1989, we analyzed 89 red king crab samples of unknown origin. These samples were from a boatload of crabs allegedly caught near Adak Island in the Aleutian Islands. Enforcement personnel from Alaska Department of Public Safety and biologists from Alaska Department of Fish and Game believed that the crabs were actually caught in Bristol Bay during an area closure. Our data clearly showed that the crabs could not have come from Adak Island and that they probably originated from the Norton Sound/Bristol Bay stock. Based on these findings the vessel owner and the skipper agreed to pay the state $565,000 in penalties for fishing violations.

We believe that these data will be of considerable use in the management of harvest of Alaskan red king crab stocks. Additionally, the knowledge by fishermen that unknown samples may be identified to stock of origin may deter illegal fishing and improve the quality of catch statistics used to manage crab fisheries.

PDF iconAdobe Acrobat PDF files require a free viewer available directly from Adobe. Need a hard copy of a publication? Contact Genetics Webmaster.