A Big Step Into the Future: CRISPR Cas9 and AI


The number of people who think that robots that have taken over the world and people with superior physical and cognitive abilities, which are frequently encountered in movies, are completely or largely science fiction is not a small number. However, groundbreaking discoveries in life and computer science in recent years can make this science fiction is real. In this direction, in this article, the most popular of artificial intelligence and genetic engineering techniques, the CRISPR Cas9 technique, which has recently made great progress in their fields, will be examined and the possible consequences of their combination will be discussed.

Genetic Engineering 101

Genetic engineering (GE) is a field that includes techniques that aim to change DNA for purpose. The main purpose in this field is to change, improve and even add new features to living things by interfering with hereditary characteristics. There are several techniques used for these interventions. The best known of these are Zinc Finger Nuclease, TALENs and CRISPR Cas9 system.

Road to the Crispr Cas9

Although there were many techniques used in GE prior to CRISPR Cas9, the leading ones include Zinc Finger Nuclease (ZFN) and TALEN. Zinc finger nucleases (Figure 1) represent the first stage of effectively targeted nucleases (1). Zinc finger nucleases, which recognize two neighboring regions, divide the DNA into two with Fokl Nuclease.

Figure 1: Zinc finger nucleases working principle. (1)

Figure 1: Zinc finger nucleases working principle. (1)

Although ZFNs are the first to targeting, they have some disadvantages. The most important of these is the synthesis period that lasts for months. In addition, other disadvantages are that it has a non-modular composition and it is not possible to design a ZFN suitable for each genomic locus.

TALENs, which were started to be used in 2011 (Figure 2), were formed with TAL effector DNA binding proteins combined with Fokl endonuclease (1). The fact that even a novice user can do it within a few days is a very advantageous factor for TALEN.

Figure 2: TALEN working principle. (1)

Figure 2: TALEN working principle. (1)


CRISPR Cas9 technology (Figure 3), which came into our lives thanks to a project that aims to investigate how bacteria fight virus infections, is a system that is constantly used by bacteria (2).

CRISPR stands for Clustered Regularly Interspaced Short Palindromic Repeats. When viruses infect a cell, they inject a DNA into it. Bacteria, on the other hand, provide an immunity gain that is transmitted between generations by adding this DNA to the bacterial chromosome through the CRISPR Cas9 system. When these pieces are added to the chromosome, the cell produces a small molecule that is an exact copy of it (shown in orange in Figure 3).

CRISPR loci combine with the Cas 9 protein and form a complex with intracellular watchdog function. They scan intracellular DNA to find a sequence that is an exact copy of the RNA they have (copy of virus RNA). When these areas are detected (shown in blue in Figure 3), Cas9 precisely cuts and destroys the virus DNA. This protein named Cas9, which destroys the viral DNA of this system, is actually the structure that enables CRISPR technology to evolve from a relatively ordinary immune tool to a GE technique that has the potential to change the fate of GE.

Figure 3: An overview of the CRISPR Cas9 system.

Figure 3: An overview of the CRISPR Cas9 system.

In studies conducted to better understand the Cas9 protein, its functions of removing and destroying the virus DNA can be used to add and remove specific DNA fragments very precisely. The process of directing the Cas9 protein mentioned here to specific DNA fragments is done through an inheritance molecule called Guide RNN (gRNA) (1).

Figure 4: 3D Crystal structure formed by Cas9 endonuclease enzyme with guide RNA (gRNA) and target DNA

Figure 4: 3D Crystal structure formed by Cas9 endonuclease enzyme with guide RNA (gRNA) and target DNA

Application Areas of CRISPR Cas9

According to Jennifer Doudna, one of the creators of the technology, we will be able to see clinical applications and even approved treatments in the next 10 years. CRISPR Cas9 has been tested in mice, monkeys, and many other organisms. In addition, Chinese scientists have shown that genetic alteration can be made in human embryos (4). In addition, it has been shown that the virus can emerge from a human cell implanted with the HIV virus. Moreover, CRISPR-based genome editing studies in cell culture were conducted in 2013 (5,6).

Stand on the shoulders of giants

Isaac Newton said, referring to his predecessors, “If I can see ahead of other people, it’s because I’m standing on the shoulders of these giants” actually applies to the CRISPR Cas9 technology and researchers, dating back to the Nobel chemistry prize in 2020. Although Emmanuelle Charpentier and Jennifer Doudna received the award, we can say that they stand on the shoulders of giants. In Figure 5, you can examine the table published between 1993 and 2014 and showing the development of the CRISPR Cas9 technique.

Figure 5: Road to the Nobel Prize

Figure 5: Road to the Nobel Prize


CRISPR Cas9, a breakthrough in genetic engineering, has a number of challenges and aspects that need improvement. In this section, we will examine the basic studies on the use of artificial intelligence (AI), which is a subject that has an impact like CRISPR Cas9 in its field to overcome these difficulties.

Although the CRISPR Cas9 technique is promising for the treatment of many diseases, the similarity of genomic regions makes the system challenging. This may cause off-target effects. When implementing, accurately estimating the off-target profile of gRNA is key to the success of the study. Also, maximizing targeted knockout effectiveness is crucial for results.


In order to overcome the challenges mentioned above, there were usually gRNA-based studies to reduce the off-target effectiveness of the CRISPR Cas9 system as much as possible and to increase its targeted effectiveness. In a study conducted in this direction, an gRNA library was created to improve the results of CRISPR Cas9, minimizing off-target effects and aiming to increase effects on the target (8). Various machine learning (ML) techniques have been used in this process. Among these, the combination of support vector machines (SVM) and logistic regression (LR) gave the best result. Azimuth, the algorithm developed with this study, can be accessed via Microsoft Azure. In another study, a tool called Elevation was created by using two interconnected ML techniques to predict off-target effects. Based on this, a system called crispr.ml has been developed for optimum gRNA design (9).

Genetic Engineering

There are studies in which deep learning (DL) techniques, which are a more innovative technique, are used in addition to ML techniques in the optimum design of guide RNA. Accordingly, the researchers developed the DeepCRISPR tool based on convolutional neural networks. In this study, in order to maximize the effects on the target location and minimize off-target activities, firstly, a supervised deep neural network was developed by training unlabeled gRNAs using deep unsupervised learning and then with tagged gRNAs (10). This model has advantages over models in its field. This model works by considering the epigenetic information in different cell types. In addition, although it has been trained with limited cell type data, it largely predicts well in new cell types and fully automates the identification of sequence and epigenetic features (10).


As a result, although the GE function of the CRISPR Cas9 system has been relatively newly discovered, it has already formed the basis of studies that have been successful. Its high applicability has brought its discoverers one of the fastest Nobel prizes in history since its discovery. As with every technique, this technique has difficulties and aspects that need improvement. In order to overcome these difficulties, complex systems are created with AI, the star of its field, and solutions are sought. Although the studies continue to increase, it is seen that they have focused only on the optimum design of gRNAs. In line with better understanding of CRISPR Cas9 technique and developments in AI field, it is predicted that studies on these two pioneering subjects will increase.


  1. Crispr Genom Modifikasyonları Kitabı, Ocak 2018
  2. TED Talks – (2015 – September) How CRISPR Lets Us Edit Our DNA
  3. Nishimasu H, Ran FA, Hsu PD, Konermann S, Shehata SI, Dohmae N, Ishitani R, Zhang F, Nureki O. Crystal Structure of Cas9 in Complex with Guide RNA and Target DNA. Cell. Şubat 2014
  4. Tang L, Zeng Y, Du H, Gong M, Peng J, Zhang B, Lei M, Zhao F, Wang W, Li X, Liu J. CRISPR/Cas9-mediated gene editing in human zygotes using Cas9 protein. Molecular Genetics and Genomics. Mart 2017
  5. Cong L, Ran FA, Cox D, Lin S, Barretto R, Habib N, Hsu PD, Wu X, Jiang W, Marraffini LA, Zhang F. Multiplex genome engineering using CRISPR/Cas systems. Science. Şubat 2013
  6. Mali P, Yang L, Esvelt KM, Aach J, Guell M, DiCarlo JE, Norville JE, Church GM. RNA-guided human genome engineering via Cas9. Science. Şubat 2013
  7. Eric S. Lander. The Heroes of CRISPR.  Cell. Ocak 2016
  8. Doench, J., Fusi, N., Sullender, M. et al. Optimized sgRNA design to maximize activity and minimize off-target effects of CRISPR-Cas9. Nat Biotechnol 34, 184–191 (2016).
  9. Listgarten, J., Weinstein, M., Kleinstiver, B.P. et al. Prediction of off-target activities for the end-to-end design of CRISPR guide RNAs. Nat Biomed Eng 2, 38–47 (2018).
  10. Chuai, G., Ma, H., Yan, J. et al. DeepCRISPR: optimized CRISPR guide RNA design by deep learning. Genome Biol 19, 80 (2018).