The ethical concerns raised by applications of CRISPR-Cas9 in gene drives and human germline editing



CRISPR-Cas9 is a revolutionary gene-editing technology developed in the last decade from a biogenic mechanism used by bacteria as an immune defence. With its low cost of operation, easy accessibility, and various published protocols in place that make gene editing easier, its countless applications raise various ethical concerns surrounding its use. Here I highlight how CRISPR-Cas9 could be employed in gene drives, designer babies, bioterrorism, and its various associated concerns. I will examine potential capabilities, constraints, and precautionary measures for regulating the whole gamut of CRISPR-Cas9 applications. The capacity to modify genetic sequences rapidly and effectively would be extremely beneficial to humans and our environment. For example, preventing the passing on of malicious diseases to future generations, effectively eradicating illnesses, assisting agriculture by undoing insect and weed pesticide and herbicide resistance, respectively, and yielding more accurate diagnoses and successful targeted treatments. However, the increasing risk of bioterrorism due to CRISPR-Cas9’s easy access and cheap costs could result in possible unwanted ecological effects, at the least, in addition to various other unforeseeable consequences. This demands a comprehensive review of each prospective application of CRISPR-Cas9. Lastly, I advocate for meaningful, open, and informed public discourse to examine the proper application of this ground-breaking technique in today’s and tomorrow’s worlds.


Genome editing is a collection of tools that allow researchers to manipulate an organism’s DNA. These methods enable for the insertion, excision, or alteration of genetic material at strategic sites across the genome. It has become a topic of immense interest in today’s world, especially in terms of disease prevention and therapy, as it is increasingly being used in research facilities to help explain illnesses through the use of cells and animal models. However, scientists are continuously trying to determine whether newer and newer technologies are safe for public use. Its capacity to possibly treat and prevent more severe illnesses such as heart disease and cancer is of particular significance.

CRISPR-Cas9—clustered regularly interspaced short palindromic repeats and CRISPR-associated protein 9—is the most commonly used technology for genome editing. The CRISPR-Cas9 system has piqued the curiosity of scientists all around the world because it is faster, more cost-effective, more accurate, and far more effective than other modern genome editing techniques1.

However, as CRISPR-Cas9 technologies have improved over the years to allow for even simpler employment of such techniques, their application in fields such as gene drives and human germline editing has been considered increasingly problematic. Unfortunately, there are several ethical and societal issues with the various promising applications of this approach. The most controversial concerns regarding human germline edits and gene drives are essentially the threats to human wellbeing, including the potential for unforeseen, adverse effects in clinical applications; threats to the environment, such as the risk of causing various unforeseeable mutations and unwelcome side effects that are damaging to the ecosystem; or others pertaining to the treatment or prevention of genetic illnesses, including the concern of informed consent and the threat of eugenic misuse. As a result, its utilisation poses several ethical and regulatory concerns that necessitate immediate government intervention and public discussion. Although the ethical dilemma over gene editing is not new, CRISPR-Cas9-mediated editing has given it a new edge.

In order to better understand the novel CRISPR-Cas9 technology, this review seeks to examine the ethical and sensible applications of CRISPR-mediated genetic manipulation through strict regulation, as well as extensive worldwide conversation and attention. Other groups of society’s perspectives, such as the wider populace and religious scholars, are equally as important to take into consideration as the scholarly debate among researchers, industrialists, ethicists, and legislators. Permitting CRISPR-based experimentation to proceed with sufficient rationale may be acceptable in countries with a well-established regulatory regime. Future clinical uses, however, must be carefully governed by freshly formulated and constantly adapting laws following careful analysis and sufficient discussion.


CRISPR-Cas9 is based off of a biogenic gene-editing mechanism employed by bacteria such as Streptococcus pyogenes to defend themselves from invading viruses. In response to viral infection, bacteria create CRISPR arrays by capturing tiny fragments of the virus’ DNA and incorporating them into their own in a predetermined order. These CRISPR arrays allow bacteria to “remember” these viruses, just like memory T and B lymphocytes do in humans. In the eventuality that the virus returns, the bacteria use these CRISPR arrays to synthesise RNA segments that recognise and bind to corresponding regions of the virus’ DNA. The virus’s DNA is subsequently cut apart and rendered inactive by Cas9 or a related enzyme. Its functioning is analogous to how our human immune system works, where, after initial infection or vaccination, we can “remember” the invader and fight it off more readily the next time around due to the formation of memory cells, the counterpart to CRISPR arrays.  In 2013, Researchers eventually adapted this natural immune defence system to edit DNA. Since then, the technique has gained increased traction as a result of its ability to genetically edit a variety of organisms, including both plants and animals, with unrivalled ease. The CRISPR-Cas9 system that we use today in a laboratory setting is outlined in Figure 1, where it comprises two essential molecules that create a cut in the target organism’s genome sequence. Cas9 is a “molecular scissor” enzyme that slices the double-helixed DNA at a designated site in the genome, allowing for the insertion or removal of DNA pieces. A guide RNA (gRNA) is an RNA fragment composed of a small pre-designed RNA sequence (about 20 bases long) set inside a larger RNA scaffold. The gRNA sequence (shown in red in Fig. 1) leads Cas9 to a corresponding DNA sequence once the scaffold has been successfully attached to Cas92. This assures that the Cas9 enzyme breaks the genome precisely at the intended locus. In short, every CRISPR experiment may be broken down into three key phases – plan, edit, and analyse.

Figure 1 | How CRISPR works. This figure reviews the simple and novel process for editing DNA through the CRISPR-Cas9 system. The gRNA in part 1 has multiple association domains: complementary binding in red, and association with the Cas9 in blue3

CRISPR-Cas9 can be applied to perform two antithetical procedures – gene knockout and gene insertion. Gene knockout, abbreviated as KO, is a method that includes modifying a cell’s or model organism’s genome to prevent the expression of a specific gene. Gene knockout approaches, unlike knockdown approaches, cause particular genes to be damaged and rendered inactive4. Even if the cells or model organisms survive the knockout, they will more often than not be unable to express the functioning gene product again. In rare cases, depending on the scope of the mutation introduced, such as a premature stop codon, there could be a reversion mutation that happens by chance and leads to a fitness advantage in the population that could be selected for. However, the title “gene knockout” does not entail that an entire gene must be forcibly deleted from the genome. Instead, the start site of transcription is typically where a “frameshift mutation” results in a stop codon. The gene is knocked out since all transcription below the stop codon is irreversibly stopped. Gene insertion, by contrast, is the practice of inserting one or more genes into a DNA sequence, which has conventionally been conducted by plasmid DNA or integrated viral vectors. However, in the traditional gene insertion procedure, the site of insertion is uncontrolled, which may result in undesired off-target effects, such as the interruption of the coding region of another gene. Engineered nucleases allow for targeted therapeutic genes to be delivered into pre-determined loci in the genome, such as a genomic “safe harbour,” which might reduce the danger of insertional mutagenesis and increase efficiency in a site-specific way to permit high levels of gene expression5.

These edits are carried out by introducing a splice or a double-strand break (DSB) in target sites to trigger two DNA repair mechanisms: non-homologous end-joining (NHEJ) and homology-directed repair (HDR). The consequences of these DNA repair pathways are summarised in Figure 2. Fundamentally, the ends of an NHEJ-break do not require a homologous template to be ligated, whereas HDR-breaks necessitate the use of a template to facilitate repair. During double-stranded break (DSB) repair, the NHEJ pathway induces insertions and deletions (indels) in a highly fallible process that joins broken ends of DNA, resulting in an assorted pool of indels. In contrast, HDR is regarded as the leading mechanism for accurate DSB repair. However, it is particularly inefficient due to the increased sequence similarity required between the cut and uncut donor strands of DNA. The more alike the DNA template used for repair is to the original unaltered DNA sequence, the fewer mistakes or mutations there are6. To emphasise, NHEJ leads to insertions and deletions, leading to gene knockout, while HDR allows for gene insertion.

Figure 2 | Double-strand break (DSB) repair pathways. DSBs are introduced at select locations by DNA gene-editing nucleases (TALENs, ZFNs, and CRISPR/Cas9). The presence of donor template allows for the possibility of DSB reparation by NHEJ or HDR. Indels are generated as a result of gene disruption induced by NHEJ (on the left) targeting the locus (insertion or deletion). The intermediate areas are deleted when the opposite ends of a pathogenic amplification or insertion are attacked by two DSBs resulting in a Non-homologous End Joining (NHEJ) gene alteration. A DSB is produced at the targeted locus as a result of HDR gene correction or insertion in company with a donor-corrected HDR template7. (HDR on the right).

CRISPR-Cas9’s unprecedented ease of use is further amplified by our reality of a digital world. Today, countless algorithms are conveniently accessible online to help deduce whether off-target edits in the genome sequence help prevent such off-target edits that lead to mutations in other genes besides the target of interest, the result of which could be a higher rate of cell death or a higher likelihood of transformation into a carcinogenic cell type. BLAST—Basic Local Alignment Search Tool—is one such digital method accessible for use online at the National Centre for Biotechnology Information (NCBI) website, among other easily accessible sites today. BLAST computes the statistical significance of correlations between nucleotide or protein sequences and sequence databases. Its ability to swiftly match and contrast a query DNA sequence to a database of sequences using a heuristic framework that closely simulates the Smith-Waterman algorithm makes it an indispensable tool in present genomic research8. Tools like BLAST and other massive sequencing projects have increasingly encouraged researchers today to examine the “genetic blueprint” of life among a variety of organisms across various ecosystems and have also aided the promising discipline of bioinformatics in bridging biology and computer science. Most importantly, though, tools like BLAST allow you to make sure there are no off-target sequences recognised by your gRNA of interest.

CRISPR-Cas9 systems are certainly the latest and most exciting developments in genome editing technology with their improved efficiency, availability, and low cost of use. However, genome editing has long been of great importance and popularity since its first practical uses, dating back to the 1950s. The first “genome editing” nucleases to arise were zinc finger nucleases (ZFNs), created when scientists fused the “zinc finger” motif, the most prevalent DNA binding mechanism in eukaryotes, to a nuclease. Because the nucleases with which they are associated only operate as dimers, targeting a particular locus requires two ZFNs: one that reads the sequence upstream and another that reads the sequence downstream. Transcription activator-like effector nucleases (TALENs) are akin to ZFNs in that they employ DNA binding motifs to trigger the matching non-specific nuclease to cut a gene at its unique locus, but each domain recognises a single nucleotide instead of DNA triplets, thereby greatly reducing the complexity of the process. Unlike their predecessors, both techniques have the advantage of not being confined to mutagenesis in only embryonic stem cells9.
However, in comparison to the ZFN and TALEN alteration systems, the CRISPR-Cas9 method has three clear advantages:

  1. Simplicity: Because gRNAs are based on the synthesis of ribonucleotide compounds rather than protein/DNA recognition as in ZFNs and TALENs, they may be customised to target effectively any section of an organism’s genomic sequence.
  2. Efficiency: RNAs expressing the Cas protein and gRNA may be directly injected into developing embryos to introduce alterations, thereby reducing procedure time.
  3. Multiplex Alterations: By injecting several gRNAs into the same cell, alterations may be introduced in various genes simultaneously.

Compared to relatively dated gene targeting methods, all three technologies provided scientists and researchers with novel approaches that have greatly advanced the art of performing genomic edits.

Figure 3 | Integration of tracrRNA. In type II CRISPR/Cas systems, Cas9 is directed by a two-RNA structure made when tracrRNA is activated and crRNA is targeted to cut site-distinct dsDNA. (top). By attaching the 3′ end of crRNA to the 5′ end of tracrRNA, a chimeric RNA is generated10 (bottom).

In a study published in early 2021, researchers revealed a breakthrough CRISPR-based technology namely “CRISPRoff”11 that essentially permits scientists to switch off practically any gene within human cells without changing the genetic coding at all. Alongside this, they also outline the corresponding “CRISPRon” technology which can activate gene expression of a particular gene as well!


What are Gene Drives?

Molecular biology has made significant advancements throughout the decades but has frequently fallen short of addressing important biological problems that threaten human health and the environment. Even modifying the genomes of experimental animals was extremely challenging until recently12. Natural selection has greatly constrained our ability to modify ecosystems as it further negates genetically altered features because they are commonly linked with reduced evolutionary fitness.

Gene drives have the potential to alleviate ecological issues by modifying whole populations of living organisms. However, due to technical limitations, their use in the pre-CRISPR era has remained predominantly theoretical. A gene drive is a unique gene engineering mechanism wherein genes are altered to bend our laws of inheritance. They significantly increase the likelihood that a particular set of genes will be transmitted over generations, enabling the genes to rapidly diffuse throughout a population and flout Gregor Mendel’s predictions of segregation and independent assortment. It is crucial to note that gene drives are naturally occurring phenomena which have been modified to benefit humanity with minor improvements. This ability helps tackle an array of leading global concerns such as the emergence of “superbugs” and “superweeds”from increased herbicide and pesticide resistance, the dissemination of malaria and other vector-borne diseases, and various other detrimental ecological impacts. 

In recent times, these gene-editing drives are becoming easier to implement with the help of CRISPR-Cas9. Using the Cas9 nuclease to build RNA-guided gene drives is a reasonable solution to overcome the targeting and stability issues that have stymied gene drive development. However, the extent to which Cas9’s particular capabilities are well suited to overcome other molecular obstacles for the construction of safe and functioning gene drives is less evident. Apart from the ability to genetically alter insect populations to alleviate disease transmission, the advancement of CRISPR-Cas9 systems would also unearth a novel method to ecological engineering with applications ranging from human well-being to the agricultural sector12.

The three fundamental components that make up a gene drive are the gene you wish to propagate, the Cas9 DNA-cutting enzyme, and the CRISPR gRNA. It should be noted, however, that this would only work in diploid organisms – organisms with two copies of each chromosome. Many microbes and plants have different numbers of chromosomes in each cell, which typically makes it more difficult for gene drives to be effectively deployed (even though plants can perform sexual reproduction). In both chromosomes of the target organism, the genetic material that encodes for those three pieces is introduced in lieu of the naturally existing gene you would want to replace. In the case of regular inheritance, each gene has a 50% likelihood of being transmitted from parent to progeny, and gene drive techniques makes that 50% chance a near-certainty13. This increased rate of inheritance is outlined in Figure 4, where over the course of three generations, there is an exponential increase in the number of organisms harbouring the gene drive chromosome.

Figure 4 | Transmission of endonuclease gene drives. . (A) A wild-type organism (grey) mates with an organism that carries the endonuclease gene drive (blue). The gene drive is preferentially acquired by all offspring. Regardless of the damage to the organism, the edit promotes its transmission until the entire population expresses it. (B) Because the corresponding wild-type chromosome was destroyed by the endonuclease, all progeny inherited the gene-drive modification. The chromosome functions as the “repair blueprint” that reproduces the drive onto the wild-type chromosome when the cell “repairs” its break via homologous recombination. When such a repair is expected, effective gene drives must be cut on a regular basis12.

Interestingly, there has yet to be any publicised engineered endonuclease gene drive capable of replicating entirely across a wildlife population. Furthermore, using outdated gene-editing technologies like ZFNs and TALENs, given their high operational costs and other drawbacks, scientists found it far more difficult to initiate gene drives. Owing to the evolutionary fragility of the periodic repeats in such proteins, cutting and homing both gradually lose their effectiveness with time. Cutting and homing eventually lose their efficiency over time due to the inherent fragility of the periodic repetitions in such proteins. These preliminary experiments demonstrate the construction of artificial gene drives but also highlight the need to eliminate any targeted gene and preserve its stability throughout replication. The recently discovered Cas9 nuclease may offer a feasible approach.

Over the years, there has been extensive research on how gene drives may be safely and effectively administered to mosquito populations14. However, there has been very limited, if any, research on how gene drives may be successfully employed in other, particularly pest, species. It was simply impractical to create such drives to transmit a specified genetic mutation in other species using dated gene editing technologies like ZFNs. Unfortunately, today’s outlined gene drive schemes are no different as the majority run the risk of harming or completely eradicating vulnerable species thereby damaging entire ecosystems. This underscores the pressing call to refine the promising method for safe and efficient future applications. A simple online search for “gene drive” reveals the varied viewpoints on this technology and the concerns of if and when it should be used.

In the remainder of this section, I offer a brief rundown of Cas9-mediated gene editing and gene drives, and I investigate the potential of gene drives on Anopheles mosquitoes and their likely capabilities and limitations. To better comprehend how gene drive technology works, I also discuss current innovative gene drive configurations that might significantly increase our influence over gene drives and their effects, in addition to exploring potential uses and proposing standards for the safe development and assessment of this exciting yet untapped technique. 

Endonuclease Gene Drives

In order to demonstrate drive, natural homing endonuclease genes cut the appropriate site on chromosomes in which they are absent. The cell responds by homologously recombining the gene onto the damaged chromosome (Figure 4A). This copying mechanism is known as homing, and the cassette with the endonuclease that is duplicated is called a “gene drive” or just a “drive.” When genes essential for viability or fertility are targeted, homing endonuclease gene (HEG)-based gene drives can induce population suppression. Because replicating increases the proportion of children that inherit the cassette (Figure 4B), these genes will spread across a population even though they lower the fertility of the species that they are found in. This self-subsisting technique could potentially enable a gene drive to extend from a select group of people to all members in a community over several generations. Despite that, these efforts are subject to failure due to processes that produce cleavage-resistant alleles with wild-type gene activity12.

Natural Gene Drives

Specific genes ‘drive’ themselves across populations in nature by greatly increasing the likelihood of being transmitted down generations15. This selfish manipulation of procreation by genetic components to transmit themselves at a high rate to the following generations, allowing them to advance extensively through populations, is a fascinating naturally occurring phenomenon.

Engineered Gene Drives

An endonuclease recombinant gene must be introduced in lieu of the innate sequence that it functions to cleave to produce a successful gene drive. It will spread across vulnerable wild-type populations only if it meets three basic level of requirement: it successfully alters this sequence in species with a native site and an isolated recombinant gene, it avoids being expensive for the organism itself and it accurately induces the cell to replicate the recombinant gene12.

Engineered gene drives may be very helpful in treating vector-borne illnesses such as malaria because they can increase the selective inheritance of certain genes, transmit desired genes across wild populations, or suppress harmful species. Standard drives propagate genetic alterations and characteristics throughout populations. The duplicating step of this drive can occur immediately after fertilisation or even in germline cells that are direct progenitors to reproductive cells (sperm and egg), leaving most somatic cells with just a single duplicate drive.

Suppression drives diminish the strength of the populations being inhibited. When infrequent, these ‘genetic load’ drives propagate quickly among marginally compromised heterozygotes, leading the population to plummet or perhaps die out as a result of the accrued burden of recessive mutations. Another approach would be to mimic natural “gametic” or “meiotic” drives that distort the true sex ratio16.

The Malaria Gene Drive

Insect-borne illnesses have had a startling human toll over the centuries. Malaria is an acute illness spread via the bite of an infected female mosquito. A diseased mosquito may inject a few Plasmodium parasites into the human host’s vasculature when she bites for blood, and these are enough to induce infection, providing additional mosquitoes with the highly likely option to collect parasites in subsequent bites and continue the vicious cycle. Only mosquitoes belonging to the Anopheles genus, out of the 3500 or so species that exist, are responsible for transmitting human malaria. And only about 40 of those 3500 mosquitoes are responsible for transmitting malaria at a level that raises various public health concerns. Every year, it kills approximately 620,000 people, the vast majority of whom are children under the age of five, and infects another 240 million with incapacitating fevers17.

The intentional release of individuals carrying a particular advantageous genetic characteristic to spread this trait across the existing population through mating is referred to as the genetic control of insects. Given that breeding is a species-specific activity that relies on these released species reaching their target group, this strategy can have a very targeted effect. The sterile insect method, initially developed in the 1950s, is the most frequently applied mode of genetic control. The strategy depends on the mass upbringing of sterile males, which overwhelm the natural population in the area and compete for female partners with wild-type males.

Alternately, the homing endonuclease gene (HEG) drive method mentioned earlier in the paper has proven to be the most productive to date. Engineered gene drive technologies that use site-specific endonucleases to transmit preferred characteristics throughout a population were initially suggested over ten years ago, based on the activity of these homing endonuclease genes (HEGs), a class of intrinsically self-serving genetic components found in many single-cell organisms. A 15- to 30-base pair (bp) nucleotide sequence can be recognised and cleaved by HEG-encoded proteins. Homing endonuclease genes are situated inside the DNA recognition sequence, preventing further cleavage. When the HEG recognises a chromosome with the continuous recognition site, the double-strand break (DSB) created by the cut is regularly repaired using the homologous chromosome as a template, turning a heterozygote into a homozygote in a procedure referred to as “homing”18.

With this approach, the prevalence of a HEG can increase quickly within a target community. As HEGs can indeed be re-engineered to detect mosquito genes, they can theoretically be repurposed to act as an effective gene drive system in mosquitoes. A HEG established in the male mosquito germline that identifies a synthetically inserted recognition site has a higher rate of super-Mendelian heredity and, therefore, rapidly infects a confined community. Endonuclease-based gene drive systems’ greatly improved transmission rate should presumably balance the fitness expenses associated with cleavage activity and interruption of the selected locations. If this criterion is fulfilled, a drive construct will transmit across entire populations until it achieves an optimum frequency, reducing the population’s average fitness.

CRISPR-based gene drives are self-replicating genetic features used to alter whole populations of malaria mosquitoes for long-term biological control. These components, first described in 2003, employ a cut-and-paste (homing) mechanism described above in the germline to promote their autonomous proliferation from a modest starting release frequency. One potentially effective technique tries to lower mosquito populations by introducing a mutation that prevents female reproduction. To be effective in the control and eradication of this illness in Sub-Saharan Africa in particular, such mutant variations must be able to successfully challenge natural populations of Anopheles gambiae and stay functional over the long term19.

They have been observed to exhibit skewed inheritance rates approaching 100% in both Anopheles stephensi and Anopheles gambiae mosquitoes20. In principle, there are two main techniques to link a specific desired trait to a gene drive: either the gene drive “knocks out” a mosquito’s essential gene or is firmly bound to a particular effector gene (a cargo). As the gene drive transmits across the population, heterozygous organisms are highly viable, implying that they not only retain but also transmit the gene drive. In the first case, the gene drive would result in a recessive phenotypic mutation, serving as a genetic parasite. Individuals that are homozygous are not viable, reducing the population’s ability to procreate. Examples of this sort of drive have been proven within laboratory settings to have a substantial inhibitory impact on populations and have targeted genes especially crucial for female fertility and viability19.

If we were to eradicate a population of mosquitoes, Cas9 spread could principally accomplish this in two ways: it could be a gene that sensitises the mosquito to an insecticide, or it could be a gene that confers resistance to malaria within the mosquito to fight off the plasmodium parasite. Gene drives aim to address a variety of problems in fields like environmental sustainability, public health, and agriculture (Figure 5). The most significant of these is restricting the spread of infectious vector-borne diseases.

Figure 5 | RNA-guided gene drive applications. From left to right, clockwise. Malaria vectors, for example, may be designed to resist disease entry or eradicated via a suppression drive. Cas9 conveyed by a gene drive might be used to immunise wild populations that act as human virus repositories. Reversal and vaccination campaigns might aid in ensuring that all transgenes are harmless and under check. Drives may rapidly disseminate defensive genes among vulnerable or soon-to-be-endangered species, like frogs threatened by the spread of the chytrid fungus21. Locally, invasive species can be managed or destroyed without impacting other species. Sensitization campaigns might increase insecticide and herbicide environmental viability and safety. Gene drives could perhaps put ecological theories about genetic drift, speciation, sex ratios, and evolution to the test12.

Ethical Concerns and Limitations

Gene drive technologies (GDT) have sparked a great deal of scholarly debate since their preliminary trials. The security and safety of experimental research with GDTs, as well as the possible detrimental impacts on ecosystems owing to unforeseen outcomes or abuse of the technology, are a few of the cardinal concerns in this discussion. Over the years, various scholarly papers have analysed the assorted ethical implications of GDTs and have successfully mapped their “ethical landscape.”

The chief argument concerning GDTs is over whether—and if so, under what constraints—they should be used. Stakeholders on one side of the debate argue that these technologies are highly hazardous and/or ethically unacceptable on other grounds, and advocate for an interdiction on GDT field uses. Although the perspectives of unique organisations and stakeholders, besides the wide scope of moral and management issues regarding GDTs, have been acknowledged and recognised globally, the indictments that inform the standpoints of this diverse group of GDT experts have yet to be thoroughly investigated. Others, in contrast, emphasise the promise of GDTs and claim that this offers a more-than-compelling reason to further pursue these technologies. The majority of these organisations and parties support a graduated testing method in which GDTs are explored in stages, first with laboratory research, then small-scale, limited field tests, unrestricted small-scale releases, and lastly, comprehensive field releases22.

Both sides nonetheless concur that it is imperative that the technology be used in a way that is socially responsible, transparent, accountable, and consistent with regional jurisdiction and regulatory frameworks. 

Transgenic organisms engineered in laboratories act as vectors of the desired gene. When released into the wild, these transgenic organisms reproduce, and the desired gene is transmitted across the population over generations—the gene drive is put into action. The time it takes the drive to be transmitted across the entire population depends on a number of key factors besides the drive’s effectiveness. It will, however, always take several generations because the gene can only be multiplied twice in each generation. Drives will therefore spread quickly among animals that reproduce quickly and slowly among species that reproduce slowly.

Second, on an evolutionary timeline, drive-mediated genomic modifications are not everlasting. Gene drives can effectively transmit desirable traits throughout communities even if they are harmful to each individual. When the drive has hit fixation, these exceedingly detrimental features will be outcompeted by better-adapted alleles, an example of Darwin’s Natural Selection23.

Third, gene drives have no impact on animals that rely only on asexual self-fertilisation or clonal division for reproduction. All viruses and the majority of bacteria, as well as many prokaryotes, fall under this group. In populations that combine asexual and sexual reproduction, such as those of plants, standard drives that are very effective may be capable of spreading gradually. However, drives designed to repress the population will in all likelihood coerce target species to reproduce asexually to escape suppression and disallow the passing on of the selected trait. 

Given the risk of gene drives modifying whole wild populations and, in turn, ecosystems, significant safety precautions and countermeasures must be included in the technology’s development. RNA-guided gene drives possess the power to undo genomic changes that have already swept across populations. Consider a gene drive that has unintended consequences for an ecosystem or is used without permission. A later ‘reversal’ drive could undo one or all of the genetic alterations spread by the initial drive. To prevent the first drive from interfering with the reverse drive’s new sequence, it must be re-coded to the original. However, any amino acid modifications made by the initial drive might even be reversed. One such vital preventive tactic is to reverse genetic alterations at the speed of the drive rather than just having a drive-resistant allele.

By addressing specific genes or structural polymorphisms, RNA-guided gene drives can be restricted to a particular target species or subpopulation that is genetically unique by focusing on particular genes or structural variations. These ‘precision drives’ won’t be capable of spreading across non-target populations unless the sequence is substantially similar, as they can only cut a single sequence12

Alternatively, rewriting sequences targeted by the undesirable drive through an “immunising” drive might prohibit said drive from being cloned. This might be done before or in response to an unwanted drive, , and it would progress in a manner analogous to the undesirable drive. By spreading across both wild-type and gene-drive-affected animals, a combined immunising reversal drive can change both types into a modified variant that cannot be overrun by the undesirable drive. This might be the easiest method to stop an operational gene drive. Similar to conventional reversal drives, however, any ecological degradation cannot be reversed.

All studies in this promising area of research must be conducted in complete transparency, with unbiased expert evaluations of potential consequences, and through deliberate and truly inclusive public discourse that occurs only after a thorough and open evaluation procedure. This is extremely important because any consequences of introducing RNA-guided gene drives into ecosystems would be borne by the local community, if not the entire world. It is strongly urged that all labs working on conventional gene drives that can transmit across wild populations concurrently design reversal drives capable of restoring the original native phenotype. Likewise, powerful suppression drives that lead to the eradication of entire species should be built in conjunction with an immunising drive. These procedures would enable the implications of an unintentional release to be promptly mitigated, albeit only partially. Environmental sample amplification or metagenomic sequencing12 could perhaps be used in theory to track the abundance of gene drives in the wild, though more research on potential surveillance systems will be necessary.

In conclusion, choosing to use a specific drive should always be done following a thorough cost-benefit analysis on the basis of its unique function. In other words, a drive’s most feasible outcomes, such as the effects it may have on non-target organisms or on the environment and our society as a whole, among other factors, must be individually and deliberately assessed to justify its implementation.

Given today’s rates of technological advancement in Cas9 and the various results attainable through employing even the most basic gene drives, it is increasingly obvious that the growing number of ethical concerns show no signs of slackening. Hence, more and more responsibilities fall into the hands of tomorrow’s ethicists and lawmakers.


Germline vs Somatic Cell Editing

Somatic cell editing is a cutting-edge technique used by today’s scientists to modify disease-causing DNA in non-reproductive body cells. By restricting the modifications to these somatic cells alone, by definition, these edits cannot pass down to future generations.  Thereby resolving the primary ethical concern with human germline editing of random damaging mutations involuntarily being passed on to future generations.

The extensive applications of the CRISPR-Cas9 system within the human body have proven to be a promising therapy platform for a variety of malignant illnesses and conditions. Its application in treating sickle cell disease is particularly noteworthy. Recently, the therapy has been successfully applied to young patients to prevent the irreversible complications of the disease. Sickle cell disease (SCD) is a monogenic, autosomal recessive condition impacting the protein haemoglobin found in red blood cells, which is responsible for the transport of oxygen in our bodies. Affected patients’ red blood cells are distorted into a “sickle” shape by the mutant haemoglobin S molecules. The rapid cellular deterioration of these red blood cells can cause anaemia, commonly resulting in exhaustion, shortness of breath, and delayed development in children. High blood pressure in the major arteries that feed the lungs (pulmonary hypertension) can even result in heart failure. Until lately, the only viable cure for this Mendelian disease was a bone marrow transplant, but CRISPR gene therapy has granted patients a newly renewed hope.

Sickle cell disease develops due to a base-pair mutation in the ß-globin gene’s DNA sequence (HBB). A bone marrow transplant from a healthy and compatible donor was the only cure for sickle cell anaemia and SCD in general until recently. Some of the major drawbacks of this approach are locating an appropriate donor, immune transplant rejection, and graft-versus-host disease (GVHD). Since related dangers of SCD  rise with the patient’s age, most patients do not live till adulthood. Caused by a genetic abnormality, SCD is an ideal candidate for CRISPR-mediated gene therapy. The most popular CRISPR-mediated treatment is via an ex vivo process called gene-edited cell therapy, in which the patient’s hematopoietic stem cells are harvested from the bone marrow cells that manufacture all of the body’s red blood cells, repaired, and then reintroduced into the patient. The CRISPR-Cas9 nuclease targets the malfunctioning region of the ß-globin gene and is being used in recent trials to stimulate sickle mutation repair by replacing the abnormal DNA segment with the normal one. Electrical pulses are used to produce holes in the membranes of the patient’s blood stem cells through which the CRISPR-Cas9 system enters the stem cells and goes to the nuclei to rectify the sickle cell mutation24. These edited cells can then be reintroduced back into the patient, allowing them to produce healthy red blood cells without any of the risks or side effects that would come from a bone marrow transplant such as GVHD.

These encouraging results demonstrate how the CRISPR-Cas9 technology has revolutionised modern medicine and, in turn, our future as a whole. On the other hand, human germline editing introduces genetic alterations to germ cells, embryonic precursors of the gametes, or reproductive cells—oocytes and sperm. Early-stage embryos are edited in a laboratory where individual genomes are modified and inherited by generations to come.

For obvious reasons of safety, ethics, and social equity, however, the general population and international scientific community firmly concur that germline engineering in humans is a red line which should not be breached. The key differences between somatic cell and germline genetic editing are outlined in Figure 6. Today, more than 40 nations have laws forbidding germline engineering for procreation, and the Council of Europe has a legal international convention prohibiting it. Though countries like China, in particular, have had an infamous history of criminally testing germline edits over the years, all of which spotlighted novel ethical concerns that demanded immediate legal resolution. He Jiankui, a Chinese biophysics researcher, claimed in November 2018 that he had developed the first human genetically altered infants. Expecting to be showered with awards and laureates, Jiankui’s announcement was instead met with widespread public outrage, which prompted his arrest in December 2019 after careful ethical deliberation by a team of Chinese lawyers. He modified DNA in human embryos using the CRISPR-Cas9 technique to make “Lulu and Nana” less vulnerable to HIV. The modifications were introduced to restrict CCR5, a gene that codes for a protein that permits HIV to infect a cell. Scientists from all around the globe criticised his efforts, claiming that gene-editing technologies were too nascent to be used for reproductive aims. Experts also said that the experiment itself was inappropriate because it carried the risk of inducing a mutation with possibly fatal implications while yielding, in return, little benefit—reducing the offspring’s risk of developing HIV. In light of the international debacle, researchers demanded a prohibition on genome engineering in embryos and germline cells. Members of the global scientific community voiced their concerns about the extent of consent He had acquired from the newborn’s parents and the degree of transparency regarding the gene edits25. Modern germline edits follow the same scheme as outlined above in the CRISPR-Cas9 introductory section.

Figure 6 | Somatic vs. Germline Gene Editing. The editing process, risks, and consensus of somatic and germline gene editing by virtue of their principal ethical concerns26.

Ethical Concerns

The possible hazards of CRISPR-Cas9 use have provoked intense legal and ethical normative debate, despite growing expectations and aspirations for the effective and reliable treatment of acute hereditary human disorders thought to be incurable. Its ethical evaluation is broadly divided into two categories27:

Those stemming from its potential failure.

Individuals should not be exposed to the medical implications of therapies with certain adverse impacts if the risks outweigh the benefits. The gravity of the risks of off-target or unanticipated effects in human germline editing is still to be quantified. As a result, precautions against erroneous or premature efforts at this intervention should rely on current systems controlling the clinical introduction of other reproductive medicines, at the very least.

CRISPR-Cas lacks sufficient precision, resulting in off-target effects or unforeseen mutations in non-target regions of the genome with uncertain repercussions for treated cells. The frequency of off-target edits largely depends on which part of the genome is being targeted. Most gene-editing proteins have only ever been tested in human or mouse cells cultivated in culture, rarely in human embryos resulting in a viable pregnancy.

The incidence of these inaccuracies presumably varies between human and mouse cells, as well as between embryos and mature cells. Nevertheless, the quantity of inaccuracies will seldom be zero. But it is theoretically impossible to have a 100% success rate in medicine. Keep in mind that there are permissible off-target effects for many commercially available drugs—Tylenol has a risk of overdose and liver damage, and Ibuprofen can cause ulcers with chronic use, but we do not expect perfection 100% of the time with no exceptions. Since these genetic treatments are longer-lasting than ordinary pharmaceutical drugs and irreversible, we need to strive to be as close to perfection as possible to avoid any potentially grave off-target mutations.

However, on-target but unintended DNA modifications may be a larger concern than off-target consequences. It is up to the cell to repair the “wound” once Cas9 or an equivalent enzyme cuts the DNA. But, each cell’s healing mechanisms, such as non-homologous end joining or homology-directed repair, are hard to control.

Researchers have recently devised methods to circumvent the challenges posed by unpredictable DNA repair mechanisms. Strecker et al. (2019)28 describe a CRISPR technique that integrates DNA into the genome without disrupting both strands can avoid the need for DNA repair. Base editing is another technique, which entails combining a Cas9 protein that is inhibited with an enzyme that can change one DNA letter into another. The weakened Cas9 directs the base editor to its intended gene destination, where it chemically edits the DNA without causing a break. However, the likelihood of off-target alterations in base edits is also not zero.

Lastly, genes can vary between both members of a population and between individual cells. The advent of low-cost, quick genome sequencing of CRISPR-modified cells has revealed that mosaicism, as it is termed, is more prevalent than previously imagined. Mosaicism often causes issues with gene editing. For example, an embryo that has been modified to repair a gene linked to Huntington’s disease may include a hybrid of uncorrected and corrected cells. Which cells were and were not modified will determine the well-being of the emerging child. It is impossible to predict ahead of time, but mosaicism poses various ethical and legal concerns29. There are a few options available today to resolve this issue. Injecting the CRISPR system into an embryo shortly after fertilisation and then eliminating it a few hours later would make certain that genetic corrections occur before DNA replication. This method has already reduced mosaicism in monkey embryos30. One approach is to fix and then use the DNA in prospective parents’ stem cells to make gamete cells with rectified DNA. Yet, with every solution, a new problem arises.

Those that arise as a result of its success.

The potential ethical consequences of these cutting-edge technologies, when and if they work as designed, however, go beyond the theoretical and thus far unidentified hazards of human germline editing.

The repercussions of such a technique on future generations whose DNA is inadvertently altered without their consent are one of the most significant challenges with germline editing. Clinical ethics acknowledges that, until children develop their own agency and ruling competence, parents are often the best medical decision-makers for their children. This is predicated on the notion that parents ultimately make decisions that have an influence on their children’s developmental values and views, from which they stand to gain or lose the most. Therefore, it can be reasonably established that parents ultimately make the best decisions for their children.

However, there have been exceptional cases, such as in Turpin v. Sortini31, where the court overturned a parent acting in their own self-interest under the Parens Patriae jurisdiction. The notion that one would have been better off if they had never been born in the first place has not been popular with the public at large or with the law, despite reports of adults and children disagreeing with a parent’s medical decisions during gestation or infancy, especially when death was a very real possibility. So-called “wrongful life” lawsuits have traditionally been rejected based on the same postulate. If a child was born following a CRISPR edit to their genome, they would likely not have any legal standing to sue their parents for “wrongful life”27. But this may be far more complicated if the child has a mutation as a direct result of an off-target CRISPR edit that shortens their lifespan, perhaps by inducing cancer. Is it possible for the parents to have behaved recklessly and harmed their child with the genetic edit? This question has no clear answer, but if a parent were persecuted for abuse or misuse of prescription drugs in their child, I’d envision a similar result in this case.

Ethical concerns regarding non-maleficence arise when considering the possibility of putting undue strain on the ensuing child and fostering familial imbalance. The capacity to simply request reasonable therapies aimed at lowering medical risks or introducing modifications to genetically “improve” their children may cause parents to be less understanding of perceived flaws or differences among their own family members. If clinical or commercial usage of human germline editing erodes the parental instinct for unconditional acceptance, it may not be in the best interests of the afflicted individual. At the very least, the risk of injury to people and families, the consequences of which we can only imagine, provides a compelling case for caution and additional study. By moving with prudence, we can ensure a deeper insight into the potential hazards and advantages of gene editing from a scientific standpoint, allowing families to exercise their autonomy more fully through a process of informed consent.

At the societal level, two key ethical challenges concerning germline editing arise: qualms about eugenics and concerns regarding equitable access to such technologies.

Eugenics encompasses selecting beneficial features (positive eugenics) and eliminating undesirable illnesses or traits (negative eugenics). Eugenics, in any form, is problematic due to its potential to promote biases and erode standards of cultural norms. This is especially important if “enhancement” extends beyond the treatment of medical conditions. In the past, eugenics has been associated with racist ideologies and distorted notions of genetic determinism, with its forceful implementation having disastrous effects. The term has long been considered taboo, having been closely associated with the most atrocious crimes of the twentieth century and the infamous Adolf Hitler.

Though it appears highly doubtful that eugenic germline editing applications will result in a reduction of genetic diversity in subsequent generations of entire populations, they may have a significant impact on specific subgroups of the population that have the capacity and willingness to impose modifications, such as in the case of Down syndrome32. Consider a future where a government mandates that people with Down’s syndrome or other disorders must undergo germline gene editing before reproducing. Who ultimately ought to make such decisions as to who gets to reproduce in society? Today, it is generally agreed that reproduction is an individual choice and right, not to be interfered with by the government. But with science’s rapidly growing progress, who’s to say what the future holds? 

These are some of the immediate objections to eugenics that the general public and the international bioethics community have voiced, but possibly the most strongly held worry is abstract: the notion that by labelling certain individuals as “unworthy,” we risk losing sight of what it means to be human as a species. A common fear is that by making such alterations that modify the germline and therefore impact future generations, we may be “overstepping” and “playing God.” This type of individual selection conveys a message about the “fitness” of such attributes or disorders and reflects on the intrinsic worth of people who have that particular trait or condition.

Lastly, if human germline editing were highly effective and clinically administered, one of the most significant and widespread consequences might be an increase in existing disparities within and across communities. Currently, the therapeutic applications of human germline genome editing are purely speculative and limited to lab research, and all talk of cost or access remains entirely theoretical. However, it is almost certain that human germline editing will be costly. If it ever becomes a reality, access will be regionally constrained and may not be funded by all private health insurers and healthcare systems. Poor access and cultural variations influencing uptake might result in significant discrepancies in the comparative incidence of a given sickness by area, socioeconomic position, or ethnic group. Genetic disorders, which were formerly a worldwide common denominator, may now become an artefact of social status, geography, and culture33.


These claims have been made in the context of germline manipulation for many years now, long before the CRISPR-Cas9 technology was discovered. It is the view of the author that, given the type and amount of outstanding ethical, scientific, and regulatory issues, practising germline gene editing that results in childbirth is undesirable at this moment. However, it is permissible to carry out additional research on gametes and embryos using CRISPR, with regular monitoring and donor consent, to support research on potential future therapeutic uses of gene editing with meticulous supervision of off-target mutation rates, as long as it does not result in a viable offspring. There should, therefore, be no limitations on using government funds to support these revolutionary studies with infinite potential.


CRISPR is undoubtedly a revolutionary gene-editing technology, with its applications ranging from editing mosquitoes to human cells. Today, its applications may be limited; tomorrow, it may be able to completely eradicate animal-borne diseases through natural, engineered, or endonuclease gene drives, or potentially cure fatal illnesses through cost-effective treatments in a matter of minutes. The possibilities are endless.

The possibly far-reaching repercussions of CRISPR-Cas9 in human germline editing and gene drives necessitate a cautious and measured solution. Numerous practical challenges must be addressed before either may handle any of the listed uses. Many of my recommendations and forecasts are most likely to fail because biological systems are far more intricate and harder to manipulate. Nonetheless, as the current rate of scientific progress associated with CRISPR-Cas9 technologies rapidly grows, so does the number of consequences of employing the most basic of gene drives or human germline edits. Living in a reality thought unimaginable just a decade ago alongside the first CRISPR-edited humans, He Jiankui’s Lulu and Nana, hints that molecular biologists may soon be able to accomplish incredible feats never imagined before.

Prospective clinical applications of CRISPR-Cas9, particularly in gene drives and human germline editing, must be deferred unless there is, at the very least, (a) a convincing medical justification, (b) an ethical rationale, (c) evidence supporting its clinical applications, and (d) an open and honest process to incorporate actionable insights and accommodate stakeholder evaluation27. By introducing these prospects to the international scientific community and the general public’s attention before they are implemented in the laboratory, I seek to initiate open, informed, inclusive, and ethical debates on the responsible assessment and deployment of these emerging technologies to prevent the potentially lethal misuse of CRISPR technologies.

  1. MedlinePlus. What are genome editing and CRISPR-Cas9?. (2022 []
  2. F. A. Ran, P. D. Hsu, J. Wright, V. Agarwala, D. A. Scott, F. Zhang. Genome engineering using the CRISPR-cas9 system. Nature Protocols. 8, 2281-2308 (2013 []
  3. How It Works. CRISPR: A potential cure for genetic diseases. (2016 []
  4. A. Ishibashi, K. Saga, Y. Hisatomi, Y. Li, Y. Kaneda, K. Nimura. A simple method using CRISPR-Cas9 to knock-out genes in murine cancerous cell lines. Scientific Reports. 10, 22345 (2020 []
  5. M. Song, Y. Kim, J. Kim, H. Kim. Chapter Five – Genome engineering in human cells. Methods in Enzymology. 546, 93-118 (2014 []
  6. Y. Miyaoka, J. R. Berman, S. B. Cooper, S. J. Mayerl, A. H. Chan, B. Zhang, A. Karl-Neumann, B. R. Conklin. Systematic quantification of HDR and NHEJ reveals effects of locus, nuclease, and cell type on genome-editing. Scientific Reports. 6, 23549 (2016 []
  7. H. Li, Y. Yang, W. Hong, M. Huang, M. Wu, X. Zhao. Applications of genome editing technology in the targeted therapy of human diseases: mechanisms, advances and prospects. Signal Transduction and Targeted Therapy. 5, 1 (2020 []
  8. I. Lobo. Basic Local Alignment Search Tool (BLAST). (2008 []
  9. J. Yeadon. Pros and cons of ZNFs, TALENs, and CRISPR/Cas. (2014 []
  10. M. Jinek, K. Chylinski, I. Fonfara, M. Hauer, J. A. Doudna, E. Charpentier. A Programmable Dual-RNA–Guided DNA Endonuclease in Adaptive Bacterial Immunity. Science. 337, 816-821 (2012 []
  11. J. K. Nuñez, J. Chen, G. C. Pommier, J. Z. Cogan, J. M. Replogle, C. Adriaens, G. N. Ramadoss, Q. Shi, K. L. Hung, A. J. Samelson, A. N. Pogson, J. Y. S. Kim, A. Chung, M. D. Leonetti, H. Y. Chang, M. Kampmann, B. E. Bernstein, V. Hovestadt, L. A. Gilbert, J. S. Weissman. Genome-wide programmable transcriptional memory by CRISPR-based epigenome editing. Cell. 184, 2503-2519 (2021 []
  12. K. M. Esvelt, A. L. Smidler, F. Catteruccia, G. M. Church. Emerging technology: Concerning RNA-guided gene drives for the alteration of wild populations. eLife. 3, e03401 (2014 [] [] [] [] [] [] [] []
  13. J. Champer, A. Buchman, O. Akbari. Cheating evolution: Engineering gene drives to manipulate the fate of wild populations. Nature Reviews Genetics. 17, 146-159 (2016 []
  14. N. Alphey, M. B. Bonsall. Interplay of population genetics and dynamics in the genetic control of mosquitoes. Journal of the Royal Society Interface. 11, 2013071 (2014 []
  15. A. Burt, R. Trivers. Genes in conflict: The biology of selfish genetic elements. []
  16. R. J. Wood, M. E. Newton. Sex-ratio distortion caused by meiotic drive in mosquitoes. The American Naturalist. 137, 379-391 (1991 []
  17. Centers for Disease Control and Prevention. Sickle Cell Disease (SCD). (2020 []
  18. A. Hammond, R. Galizi, K. Kyrou, A. Simoni, C. Siniscalchi, D. Katsanos, M. Gribble, D. Baker, E. Marois, S. Russell, A. Burt, N. Windbichler, A. Crisanti, T. Nolan. A CRISPR-Cas9 gene drive system targeting female reproduction in the malaria mosquito vector Anopheles Gambiae. Nature Biotechnology. 34, 78-83 (2016 []
  19. T. Nolan. Control of malaria-transmitting mosquitoes using gene drives. Philosophical Transactions of The Royal Society B Biological Sciences. 376, 20190803 (2020 [] []
  20. J. Champer, J. Liu, S. Y. Oh, R. Reeves, A. Luthra, N. Oakes, A. G. Clark, P. W. Messer. Reducing resistance allele formation in CRISPR gene drive. 115, 5522-5527 (2018 []
  21. E. B. Rosenblum, J. Voyles, T. J. Poorten, J. E. Stajich. The Deadly Chytrid Fungus:A Story of an Emerging Pathogen. PLOS Pathogens. 6, e1000550 (2010 []
  22. N. de Graeff, K. R. Jongsma, A. L. Bredenoord. Experts’ moral views on Gene Drive Technologies: A qualitative interview study. BMC Medical Ethics. 22, 25 (2021 []
  23. A. Kormos, G. C. Lanzaro, E. Bier, V. Santos, L. Nazaré, J. Pinto, A. A. dos Santos, A. A. James. Ethical considerations for gene drive: Challenges of Balancing Inclusion, Power and Perspectives. Frontiers in Bioengineering and Biotechnology. 10, 826727 (2022 []
  24. R. Sanders. FDA approves First Test of CRISPR to correct genetic defect causing sickle cell disease. (2021 []
  25. D. Cyranoski. What CRISPR-baby prison sentences mean for research. Nature. 577, 154-155 (2020 []
  26. M. T. Bergman. Perspectives on gene editing. (2019 []
  27. K. E. Ormond, D. P. Mortlock, D. T. Scholes, Y. Bombard, L. C. Brody, W. H. Faucett, N. A. Garrison, L. Hercher, R. Isasi, A. Middleton, K. Musunuru, D. Shriner, A. Virani, C. E. Young. Human germline genome editing. AJHG. 101, 167-176 (2017 [] [] []
  28. J. Strecker, A. Ladha, Z. Gardner, J. Schmid-Burgk, K. S. Makarova, E. V. Koonin, F. Zhang. RNA-guided DNA insertion with CRISPR-associated transposases. Science. 365, 48-53 (2019 []
  29. H. Ledford. CRISPR babies: When will the world be ready?. Nature. 570, 293-296 (2019 []
  30. M. L. Page. Mosaic problem stands in the way of gene editing embryos. (2017 []
  31. K. J. Kearl. Turpin V. Sortini: recognizing the unsupportable cause of action for wrongful life. California Law Review. 71, 1278-1297 (1983 []
  32. J. F. X. Egan, K. Smith, D. Timms, J. M. Bolnick, W. A. Campbell, P. A. Benn. Demographic differences in down syndrome livebirths in the US from 1989 to 2006. Prenatal Diagnosis. 31, 389-394 (2011 []
  33. National Academies of Sciences, Engineering, and Medicine. Human genome editing: Science, ethics, and governance. []