Bioinformatics procedure for investigating senolytic (anti‐aging) agents: A digital signal processing technique

Abstract Objective Cell growth involves cell division. This stops after reaching a certain limit. Some cells become inactive and unable to undergo apoptosis (programmed cell death). These cells accumulate at sites of tissue damage or disease, thus accelerating aging. They are called senescent cells. Therapeutic interventions that can either eliminate senescent cells (senolytics) or suppress their harmful effects (senomorphics) have been developed. Senescence (aging) is caused by the inter‐ and intramolecular interactions between the domains of forkhead (FHD) and transactivation (TAD), as well as C‐terminal region 3 (CR3) and DNA binding (DBD). On the other hand, anti‐senescent/senolytic (anti‐aging) activities are achieved by disrupting these interactions with CR3‐ and forkhead box protein O4 (FOXO4)‐based peptides, such as ES2 and DRI, respectively. In this study, we use a computerized procedure based on digital signal processing to systematically analyze the inter‐molecular interactions between senolytics and their targets. Methods Informational spectrum method (ISM) is engaged. Results We obtained the sequences of the peptides from the interacting proteins of CR3 and FOXO4 and evaluated their ability to disrupt the inter‐molecular interactions between FOXO4 and DRI and CR3 and BDB, which are responsible for senescence (aging). Our results show that the peptides have different degrees of senolytic (anti‐aging) activity, depending on their affinity for CR3 and BDB, or FOXO4 and DRI. We found that enhanced senescence 2 (ES2) has a higher affinity for CR3 and BDB than FOXO4 and DRI, and that the interaction between CR3 and BDB is crucial for aging. Therefore, ES2 and other CR3‐based peptides are more potent senolytics than FOXO4‐based peptides. Our findings are consistent with previous studies and reveal new insights into the mechanisms of senescence and senolytics. ES2 is considered the best senolytic candidate, as it is 3–7 times more effective than DRI. We verified that ES2 has a weaker interaction with FOXO4 than CR3. However, the performance of DRI has been noted to depend on its intramolecular interactions and stability. Hence, intramolecular analyses using the digital signal processing‐based technique has become very vital and will follow. Conclusion CR3‐based peptides are promising candidates for senolytic therapy. Senolytics are linear chains of amino acids that can target and eliminate senescent cells, which are cells that have stopped dividing and contribute to aging and age‐related diseases. By using this proposed, novel computerized technique that is based on digital signal processing, senolytics can be easily analyzed and optimized for their effectiveness and safety. This provides a more rational approach to enhancing our longevity and well‐being by offering interventions that can delay or reverse aging and insights that can advance the field of gerontology. This procedure also will compliment other approaches such as molecular stimulation, etc.

of CR3 and FOXO4 and evaluated their ability to disrupt the inter-molecular interactions between FOXO4 and DRI and CR3 and BDB, which are responsible for senescence (aging).Our results show that the peptides have different degrees of senolytic (anti-aging) activity, depending on their affinity for CR3 and BDB, or FOXO4 and DRI.
We found that enhanced senescence 2 (ES2) has a higher affinity for CR3 and BDB than FOXO4 and DRI, and that the interaction between CR3 and BDB is crucial for aging.Therefore, ES2 and other CR3-based peptides are more potent senolytics than FOXO4-based peptides.Our findings are consistent with previous studies and reveal new insights into the mechanisms of senescence and senolytics.ES2 is considered the best senolytic candidate, as it is 3-7 times more effective than DRI.We verified that ES2 has a weaker interaction with FOXO4 than CR3.However, the performance of DRI has been noted to depend on its intramolecular interactions and stability.Hence, intramolecular analyses using the digital signal processing-based technique has become very vital and will follow.

| INTRODUC TI ON
Cell growth can involve cell division or differentiation.However, cells can stop growing due to natural limits or external stress, such as cancer or therapy.Some cells die and are removed (apoptotic), while others remain inactive and accumulate in the body (senescent).2][3][4][5] SASP is associated with various diseases, such as diabetes, 6 osteoarthritis, 7 atherosclerosis, 8 neurodegeneration, and cancer. 9,10Senescent cells avoid death and aging by interacting between two proteins, forkhead box protein O4 (FOXO4) and p53, 2,11 which regulate cell fate.Specifically, the interaction between the C-terminal region 3 (CR3) of FOXO4 and the DNA binding (DBD) of p53 as well as D-retro-inverso (DRI) and FOXO4 is crucial for senescence. 32][3][4] Aging is known to be driven essentially by cellular senescence and other factors, 1,2,11 processes that impair cell function and growth.To explore how we can target and eliminate these harmful cells with senolytic agents, we used a bioinformatics approach to investigate the inter-molecular interactions between senolytics and their protein partners.Intra-molecular studies are also envisaged in future.Some anti-aging agents, such as Quercetin, Dasatinib, Fisetin, Navitoclax, and DRI, can clear the effects of senescent cells from our body either by removing the cells or inhibiting their activities.Those agents that remove the senescent cells are called senolytics. 1,2They include DRI and enhanced senescence 2 (ES2).DRI is a reversed sequence of a part of FOXO4 protein, fused with a cell-penetrating peptide (PPRRRQRRKKRG) from HIV-TAT. 2,11,122][13] This is recorded in Table 2.
A clinical experiment using nuclear magnetic resonance (NMR) has shown that DRI not only restores frailty, fitness, lifespan, and healthspan, it also reinstates fur density and kidney dysfunction in aged mice as well as in mice with clinically induced aging pathologies. 2,3ES2 has been identified to be 3-7 times more effective than the DRI. 12 The two sets of peptides, namely FOXO4-and CR3-based, are analyzed to re-evaluate process, validate, and reaffirm the initial findings acquired through alternative methods.Our investigation focuses on two peptidic senolytics along with eight interacting peptides, as outlined in Tables 2 and 3. Additionally, this study incorporates an examination of molecular descriptors.To substantiate the selection of these molecular descriptors within the research context, we prioritize an initial assessment of the biophysical properties governing both interand intra-molecular interactions.The interactions between forkhead box protein O4-transactivation (FOXO4-TAD) as well as C-terminal region 3-DNA binding (CR3-DBD) that contribute to senescence and, consequently, aging, encompass both inter-and intra-molecular interactions. 3Since this study specifically focuses on inter-molecular interactions, only the molecular descriptor, the electron ion interaction potential (EIIP), that governs inter-molecular interactions is employed.
Complementary information from the intra-molecular characteristics Conclusion: CR3-based peptides are promising candidates for senolytic therapy.
Senolytics are linear chains of amino acids that can target and eliminate senescent cells, which are cells that have stopped dividing and contribute to aging and agerelated diseases.By using this proposed, novel computerized technique that is based on digital signal processing, senolytics can be easily analyzed and optimized for their effectiveness and safety.This provides a more rational approach to enhancing our longevity and well-being by offering interventions that can delay or reverse aging and insights that can advance the field of gerontology.This procedure also will compliment other approaches such as molecular stimulation, etc.

K E Y W O R D S
C-terminal region 3, digital signal processing, DNA binding, D-retro-inverso, enhanced senescence 2, eterone 1, forkhead box protein O4, forkhead domain, senolytics, senomorphics will later commence.Notably, there are more than 600 molecular descriptors, based on amino acid scales, 14 that are readily available. 15her alternative approaches for deriving suitable descriptors that regulate these interactions exist.
In this study, we reevaluated senolytics and their interacting domains using a digital signal processing approach, despite both being protein/peptide-based molecules.This choice is attributed to the fact that proteins and peptides are well-recognized for their linear arrangement of amino acids, which can be viewed as discrete components. 16This discrete nature makes them highly recommendable for analysis through digital signal processing (DSP) techniques, such as the discrete Fourier transform (DFT). 17Two DSP-bioinformatics-based physio-mathematical techniques have seen widespread use in exploring bio-functionalities.They are the informational spectrum method (ISM) 18 and the resonant recognition method (RMM). 19In this study, we engaged ISM.Both RMM and ISM have played pivotal roles in uncovering the biological functions of numerous proteins, exceeding 1000 in total. 19ese proteins include those associated with influenza, 20 HIV, [21][22][23] Anthrax, 24 and Ebola. 23ISM continues to serve as the cornerstone for the development of biomedical tools and devices, such as the Computer-Aided Drug Resistance Calculator 25 and Phylogenetic tree. 26The engagement of ISM in this study serves a dual purpose-not only to reaffirm the interactions but also to deepen our comprehension of their significance in senescence.Moreover, it aims to pave the way for the creation of more effective seno-and gero-therapeutic interventions.

| ME THODS
The materials and the experimental procedure are described in this session.

| Materials
The sequences of the senolytics and their target proteins are retrieved from Uniprot Database 27 and other sources. 3,11They are as shown in Tables 2 and 3.An example of a molecular descriptor (EIIP) is also shown in Table 1.

| Experimental procedure: Informational spectrum method (ISM)
ISM is applied to these agents using EIIP is as described below.

| Step 1: Translation of the protein sequences into numerical sequence (signals)
To begin, the alphabetical codes denoting the amino acids within the protein sequences under examination are replaced with numerical values corresponding to molecular descriptors.These molecular descriptors consist of 20 values, each reflecting the degree of involvement of one of the 20 essential amino acids in the interaction, spanning from physiochemical characteristics encompassing aspects like binding affinity, hydrophobicity, amphiphilicity, and various structural features such as helicity, Alpha (α), and Beta (β) properties. 22This transformation converts the protein sequences into numerical sequences or signals.Following the DFT process principle, the numerical sequences, before undergoing decomposition with DFT, should maintain a consistent window length.Consequently, numerical sequences of proteins with shorter amino acid lengths undergo zero-padding, meaning they are augmented with zeros until they reach the desired length.
It is noteworthy that zero-padding has been established to have no impact on the decomposition's results.The degree of interaction, represented as the absolute spectrum 17 is expressed as: S a signifies absolute spectrum of a given protein; X(n) represents the DFT coefficient of the signal; X*(n) is the conjugate, and n stands for numbers 1−n/2.

| Step 4: Common informational spectrum
This refers to the pointwise multiplication 17 of all the derived FFT values of the protein's sequences investigated.This operation helps obtain one position of common interaction called consensus frequency (CF). 17mmon informational spectrum is expressed as: C a represents the absolute of the CS analysis; m = 1, 2,… M; M is the number of protein/peptide sequences engaged; ∏ is the pointwise multiplication symbol.
This position shows the highest amplitude after pointwise multiplication.The consensus frequency is represented as: Similarly, any frequency (f) of interest can be determined as: By using both the CF and frequency of interest (f), degrees of interactions by any protein at any position can be determined.The results of this experiment are recorded in the subsequent session.
Table 1 provides insights into senescence and senolytic activities, (1) contingent on the involved peptide, expressed as inter-molecular (binding) interactions attributed to the 20 essential amino acids (EIIP).Tables 2 and 3 detail the peptides under scrutiny.Table 2 outlines CR3 and CR3-based peptides along with their sequences (ES2r1, ES2r2, ES1, and ES2), while Table 3 documents FOXO4related peptides and their interacting proteins (TAD, FHD, DBD, DRI, and E1).Table 4 presents the informational spectral values for CR3related peptides, displaying their amplitudes (indicating degrees of interactions).Table 5 records the informational spectral data for FOXO4-based peptides.Table 6 offers a summary of the findings, including amino acid interaction positions/frequencies, interaction degrees (percentages) in individual plots, and their comparative contributions.It has been observed that the FOXO4-p53 interaction leads to senescence (aging) through the interaction between FHD and TAD, and these interactions can be interrupted by DRI and ES2, providing senolytic (anti-aging) activity.To enhance the efficacy of DRI, it is fused with the cell penetrating peptide (CPP).To assess the impact of CPP on senolytic activities, both the naive DRI peptide and DRI, which embodies the CPP, are evaluated.These results are presented in Figure 5.To facilitate a comparative assessment of the two peptides with unequal lengths, the shorter sequence is first zeropadded, processed, and the CF determined.As shown in Figure 5 and Table 6, DRI demonstrates 95% senolytic (anti-aging) activity through inter-molecular interaction at position 12 (f = 0.261), while DRI without CPP shows 89% activity at a similar frequency, ie, position 12 (f = 0.261).The fusion with CPP results in higher senolytic activity, justifying the inclusion of CPP.
In the design of DRI, four amino acid changes were made on E1.
To investigate the impact of these changes on inter-molecular interactions, both DRI and its variant E1 are reassessed and recorded in Figure 6.DRI and E1 exhibit amplitudes of 0.424 and 0.449 at position 12 (f = 0.261), respectively.The CIS of both DRI and E1 records 0.189.These results are displayed in Figure 6.A comparative analysis of this outcome (Table 6) reveals that E1 demonstrates greater intermolecular activity (100%) than DRI (95%).This suggests that intramolecular interactions may play a more significant role in senolytic properties, as previously suggested by H. Le.
In line with the ISM principle, sharing the same frequency indeed indicates interaction, as previously demonstrated.An analogous scenario exists in the context of HIV infection progressing to AIDS, which results from the interaction between HIV viruses and the host CD4.These proteins have been observed to share frequencies of 0.035 and 0.037, respectively. 22other noteworthy finding is that ES1, a three repeat of the cell-penetrating peptide, displays a notable propensity to interact with CR3.ES1 exhibits the highest amplitude (100%) at position 3 (f = 0.999), an interacting position closer to those of DBD (f = 0.960) and CR3 (0.125) than FHD (f = 0.257) and TAD (f = 0.245).This suggests that ES1 could be a promising candidate for disrupting the

TA B L E 5
Informational spectral values of FOXO4-based peptides.
CR3-DBD interaction, pending complementary assessments with intra-molecular interactions.The interaction between CR3 and DBD is pivotal in senescence, and its disruption holds the potential for anti-aging effects.

| DISCUSS ION
The results of this computational assessment seem to align with a clinically verified discovery that there exists an interaction between the forkhead domain and the transactivation domain (TAD) of p53.This is evident as they both share a common interaction frequency point.TAD demonstrates maximum interaction (100%) at f = 0.245, while FHD exhibits a strong interaction of 98% at f = 0.257 (Table 6).However, DRI, which is responsible

ACK N OWLED G M ENTS
We wish to appreciate the AI tool ChatGPT for helping in proofreading this research.We also thank all the creators of the virtual pictures used in the Virtual Abstract.

2. 2 . 3 |
Step 3: Informational Spectrum: Decomposition of the Signal) using Discrete Fourier Transform Fourier transform is fundamentally expressed as e symbolizes exponential; w represents omega; j is called an imaginary complex number, ie, (−1) 1/2 or √-1; dt symbolizes changes in time.Because these senolytic agents are peptides, amino acids in linear formation, they are converted into numerical sequences (signals) and decomposed with DFT17 to reveal the information embedded in them.DFT is represented as:n is the discrete time index that runs from 0 to n−1, ie, 0, 1, 2,…, n−1; k is the discrete frequency index and runs 0 to n/2, ie, 0, 1, 2,…, n/2.This is because this process provides mirror image (symmetric) characteristics that remain the product of the Discrete Fourier Transform (DFT) processing.x(k) represents m member of the numerical series where N is the length of the numerical series.X(n) stands for the coefficient of the DFT.The products of DFT application are a complex number consisting of real and imaginary, and absolute values.Absolute values are usually utilized in determining the outcome of the DFT decomposition.Absolute value is represented as: n = 1, 2…, n/2 while R and I represent the real and imaginary parts respectively; j is called an imaginary complex number, ie, (−1) 1/2 or √−1 as explained above.

Figure 1 Figure 2 ,
Figure 1 displays the informational characteristics of the DBD and CR3 (DBD data omitted due to its extensive nature).Notably, DBD exhibits its highest interaction at position 21 (f = 0.960) with an amplitude of 1.3503, while CR3 demonstrates maximum interaction at position 4 (f = 0.125) with an amplitude of 0.635.Moving to Figure 2, it highlights the informational features of FHD and TAD, showing amplitudes of 0.609 and 0.411 at positions 26 (f = 0.960) and 12 (f = 0.125), respectively.Lastly, Figure 3 presents the informational attributes of ES1 and ES2, revealing amplitudes of 0.261 and 0.269 at positions 9 (f = 0.257), respectively.

Figure 4
Figure 4 illustrates the informational spectral characteristics of ES2r1, ES2r2, and their CIS.This CIS analysis includes those of ES1 and ES2, as shown in Figure 2. The CIS demonstrates a maximum amplitude of 0.146 at position 9 (f = 0.257).

F I G U R E 1 F I G U R E 2 F I G U R E 3 5 |
for anti-senescence (senolytic or anti-aging) activity through the disruption of the FOXO4-p53 interaction, demonstrates a 100% interaction at approximately the same frequency (0.261) with both FHD and TAD.DRI, which is a d-amino acid retro-reverse sequence derived from the binding domain of the FOXO4-p53 interaction and designed by Baar et al.2,3,11 , aimed for enhanced efficacy.The retro-reversal of the sequence, coupled with fusion with the CPP, has shown in this study to significantly contribute to its anti-aging effectiveness.Furthermore, this study highlights the beneficial impact of the CPP fusion, aligning with findings obtained through alternative methods.Remarkably, ES2, although selected as the candidate, demonstrates 3-7 times greater activity than DRI.It is essential to highlight that ES2, while highly active, shows a lower affinity for FOXO4-p53.This phenomenon may be attributed to what H. Le et al. suggested: Informational characteristics of DNA binding domain (DBD) and CR3 showing maximum interaction (100%) at positions 21 (f = 0.960) and 4 (f = 0.125) with amplitudes (1.350 and 0.635), respectively.Informational characteristics of Forkhead Domain (FHD) and Transactivation Domain (TAD) showing amplitudes of 0.609 and 0.411 at positions 26 (f = 0.257) and 12 (f = 0.245), respectively.Informational characteristics of ES1 and ES2 showing amplitudes of 0.261 and 0.269 at position 9 (f = 0.257); see Table 4.that the drugability or suitability of ES2 as a drug candidate is potentially influenced by intra-molecular interactions and stability, warranting further exploration.That our findings indicate that ES1, a repeat CPP peptide, shares the highest inter-molecular interaction position with CR3 and DBD demonstrates that it holds promise as a CR3-based senolytic candidate, assuming the intra-molecular features can provide similar outcomes.It is worth noting that the interaction between CR3 and DBD, which is known to facilitate the FOXO4-p53 interaction, 3 is already acknowledged as a pivotal factor in the senescence process.CON CLUS ION Interaction between FHD and TAD, as well as CR3 and DBD, are known to contribute to senescence (aging).However, interruptions by CR3and FOXO4-based peptides such as ES2 and DRI, respectively, have been identified as providing anti-senescence (anti-aging) activities.The clearance of senescent cells using senolytics or the inhibition of their activities through senomorphics is a recognized therapy that enhances both lifespan and healthspan (longevity).The involvement of these peptides in this health management was made possible by the works of H. Le et al. and Baar et al. 1-3,11 Baar et al. introduced the d-amino F I G U R E 4 Informational characteristics of ES2r1, ES2r2, and their common informational characteristics, which includes ES1 and ES2 showing maximum interaction at positions 9 (f = 0.257), respectively.F I G U R E 5 Informational characteristics of DRI-with and DRIwithout the cell penetrating peptides (zero padded and processed) showing amplitudes of (0.424 and 0. 394) at position 12 (f = 0.261), respectively.retro-reverse sequence of the FOXO4-p53 interacting domain known as DRI, while H. Le et al. modeled CR3-based peptides like ES1, ES2, ES2r1, and ES2r2, identifying ES2 as a candidate.Understanding the interactions governing these senolytics and their interacting proteins has become crucial in designing and developing more effective senotherapeutic interventions and contributing to gero-scientific studies.This necessitates the use of simple, computerized and bioinformaticsbased approaches as provided here.As discrete structured biomolecules, peptides, which are amino acids in linear formation, are easily analyzed using digital signal processing techniques.Utilizing this bioinformatic and computerized procedure has helped improved healthspan and longevity to humanity through the development of seno-therapeutics, seno-biomarkers, and gero-scientific studies.Here, we therefore present a simple, novel technique that will expand assessments and development of senolytics.This approach will compliment other procedures like molecular simulation, etc.The validity and authenticity of the procedure rest on the technologies the technique, Fourier transform (FT), has earlier provided.Radar is a technology that detects, locates, tracks, recognizes objects like airplanes in the air.It provide flight safety.Speech detector is another device that uses voice recognition to grant authorized accesses to restricted area.These technologies are both founded on same technique employed here, Fourier Transform.17 , data mining, data processing, analysis, and presentation by Norbert Nwankwo.Analysis, design, and presentation are also carried out by Ignatius Okafor.

F I G U R E 6
Informational characteristics of DRI and E1 (DRI with 4 mutations) and their common informational characteristics showing amplitudes of (0.423965, 0.444818, and 0.188587) at position 12 (f = 0.261), respectively.The boldness signifies position of interaction.TA B L E 6 Summary of the entire findings 2: Comparative contributions.

TA B L E 1
Values for the 20 essential amino acids on account of a molecular descriptor, electron ion potential (eiip).
TA B L E 2 Sequences of CR3 and CR3-based senolytics.TPP PGT RVR AMA IYK Q SQ HMT EVV RRC PHH ERC SDS GLA PPQ HLI RVE GNL RVE YLD DRN TFR HSV VVP YEP PEV GSD CTT IHY NYM CNS SCM GGM NRR PIL IIT LEDSSGNLLGRNSFEVRCACPGRDRRTEEENLRKKGEPHHELPPGSTKRALPNNT DRI LTLKK EPA SEI AQ S ILE AYS Q NG WAN RRS GGK RPP PRRRQ RRKKRG DRI without CPP LTLKK EPA SEI AQ S ILE AYS Q NG WAN RRSGGKRP E1 GRKKR RQR RRP PPR KGG SRR RAW GNQ RYA RLI RQA IESAPEKRLTL 2.2.2 | Step 2: Zero-padding of the numerical sequences (signals) Informational spectral values of CR3-based peptides.Bold and italics presentations in the tables, etc are used to point out the positions of interactions and highlight their relevance.