Amino acids exchange in patients with CKD as obtained via exome sequencing differ radically from those of natural variants 1000 genome and genomAD databases

doi:10.21203/rs.3.rs-3344794/v1

Download PDF

Research Article

Amino acids exchange in patients with CKD as obtained via exome sequencing differ radically from those of natural variants 1000 genome and genomAD databases

https://doi.org/10.21203/rs.3.rs-3344794/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background: Nonsynonymous mutations in the coding regions of human genes are responsible for phenotypic differences between humans and for their susceptibility to genetic disease.

Methods: We performed Exome sequencing on CKD patients’ genomic DNA and put the focus in understanding the role played by the amino-acid mutation spectrum (PAM) in human chronic kidney disease CKD patients. More specifically, using SIFT algorithm, we generated the amino acids substitution on proteins and compared the PAM matrix derived from CKD patients representing the amino-acid mutational spectrum to non-disease PAM matrices representing spectra of mutual amino-acid mutation frequencies derived from 1000 genome and genomAD database.

Results: We found a strong and positive correlation in term of mutabilities of amino acids distribution in human proteome and the average distribution of amino acid mutability remains higher in genomAD dataset as compared CKD and 1000 genome. The results also show a strong correlation of mutability between the three datasets, the coefficient of correlation being: (r_{CKD vs 1kg}=0.9225_,r_{CKD vs genomAD}= 0.9431 and r_{genomAD vs 1kg} = 0.9486) as well probabilistic distribution of amino acids in human proteome between the three datasets.

Conclusion: The amino acids mutability index in CKD dataset was statistically different as compared with those in 1000 genome and genomAD datasets. There is strong positive correlation in mutabilities of amino acids distribution in human proteome and the average distribution of mutability is higher in genomAD dataset as compared CKD and 1000 genome. In the CKD dataset, Arginine remains a common product of mutability from four amino acids like: Tryptophan, Histidine, Glycine and Lysine and the spike of Arginine in blood samples should be an element to trace in diagnostic profile of CKD. The data paved a way for clinical use for amino acids exchanges in chronic kidney disease using illumine platform.

Exome Sequencing

CKD

Codons Exchanges

PAM Matrix

Amino Acids Exchanges

DNA is an adaptable and dynamic molecule. As such, the nucleotides residing inside the DNA molecule undergo through multiple permutations under an evolutional process called mutation and this process spans the entire human genome. Some of these mutations can be harmless, beneficial or deleterious to human being according to the nature of the mutation. We are invariably trained under the paradigm that all biological activities in cells are embedded in three letters code that fall under the denomination as genetic codes. The change in genetic code has long been seen as the main vectors of genetic diseases and these diseases ignite as the hurtful mutations or deleterious mutations that take over the biological system. These mutations have direct impact on amino acids sequences and it was shown that the disease-causing variants usually provoke drastic changes of amino acid physico-chemical properties of proteins such as charge, hydrophobicity and geometry arrangement [1]. Variants involved in diseases tend to cause more changes of hydrogen bond network and salt bridges as compared with harmless amino acid mutations [2, 3]. Diseases-associated variants target the more conserved sequences in genome that relatively remained unchanged far back in the phylogenic tree as observed in 1 kG database [4]. However, most of phenotypical effects and molecular cause of origin of most variants remain elusive so anticipating or predicting a single amino acid change in a specific disease will be of great medical interest. Genetic variants underlying common disease may vary in frequency across population more than random variants do in the genome, perhaps because of their exposition to epigenetics events due to natural selection [5]. The Next-generation sequencing (NGS) is instrumental in the discovery of growing number of novel candidate’s genes variants and most importantly rare variants in pharmacogenes [6].

The 1000 Genomes Project was developed to serve as a big repertoire and global reference for human genetic variation and is used to serve as a baseline for subsequent research in establishing the relationship between genotype and phenotype and ultimately identifying the genetic basis of human disease [7]. Basically, the 1 kG database serves as a reference where researchers can vet multiple SNPs resulting from exome and genome sequencing.

Besides the 1kG database, the genome aggregation database or genomAD was also developed with the goal of accumulating and harmonizing both exome and genome sequencing data from a wide variety of large-scale sequencing projects. The most contemporary data set (gnomAD v2.1) grants entry to more than 125,000 exome and 15,000 whole-genome sequences from unrelated and healthy adults sequenced as part of various disease-specific and population genetic studies. It has accumulated 15708 whole genome and 125748 of exomes. Its predecessor, the Exome Aggregation Consortium (ExAC) has been mentioned in more than 4,000 publications [8]. The genomAD includes the genomes and exomes from European, Latino African, African American, South East Asian, Ashkenazi Jewish and other population. The gnomAD papers reported 241 million small genetic variants (single nucleotide variants and short insertion/deletion variants) and 335,470 structural variants (DNA rearrangements of at least 50 base pairs), compared with 7.4 million small genetic variants identified in (ExAC). Recently, the gnomAD team has already expanded the genomic resources and has released gnomAD v3 harboring 71,702 genomes.

The Genome Aggregation Database (gnomAD) has paved the way to reveal wide range of mutational constraints affecting human protein-coding genes in a catalog predicting loss-of-function variants in 125,748 whole-exome and 15,708 whole-genome sequencing datasets [8], analysis and prediction loss-of-function variants from 125,748 human exomes and 15,708 whole genomes. It also provides a roadmap for human ‘knockout’ studies into disease biology and drug-target selection [4] and large empirical assessment that provides a reference map for disease-association studies, population genetics, and diagnostic screening [9].

One way of discovering the function of a gene is to study what happens when the gene is disrupted by a mutation. These variants often have adverse results and are usually rare. (Running a large deep of genome sequencing studies gives an opportunity to investigate the effects of these loss-of-function variants to uncover important insights into human biology and diseases [10].

With the existence of myriad of variants datasets, the establishment of novel disease-datasets becomes promising. It was in this optic that we began the study of chronic kidney failure disease (CKD) with hope to establish a nsSNPs for this disease.

Chronic kidney disease (CKD) is a major global health problem and affects approximately one in ten adults worldwide [11] Genetic and epigenetic factors contribute to this complex heterogeneous and heritable disease and the heritability threshold is estimated to be high (30–75%). Although the estimation of Mendelian disorders is supposed to be responsible for 10% of ESRD cases [12], 15–30% of patients, including those with presumably acquired forms of disease such as diabetic or hypertensive nephropathy, report a family history of renal disease. This indicates the existence of a contribution of hereditary factors across clinical categories [13].

Moreover, affected individuals within the same family frequently carry different clinical diagnoses or spectrum, suggesting that they may be sharing genetic contributions across CKD subtypes [13] and recent studies support this hypothesis [14]. For example, two variants in APOL1 (Online Mendelian Inheritance in Man reference number [OMIM]: 603743) significantly increase the risk for multiple forms of CKD, including focal segmental glomerulosclerosis (FSGS), hypertension-attributed nephropathy, and HIV-1–associated nephropathy [15]. In similar way, mutations in the type-4 collagen genes (COL4A3 [OMIM: 120070], COL4A4 [OMIM: 120131], and COL4A5 [OMIM: 303630]) traditionally associated with Alport syndrome can also manifest frequently as FSGS [16].

Collectively, these findings highlight the potential of an in depth of genetic and sequencing studies to identify the shared causative variants spanning all the spectrum of CKD and ultimately undertake clinical studies to uproot the basis of underlying molecular pathophysiological causes of CDK. (to here 7%) ok correction.

To address some part of this issue, we conducted exome sequencing on gDNA of 12 young patients suffering of kidney failure. After bioinformatics analysis of our exome sequencing data, more than 250 SNPs per sample were retained as variants of interest. Combined harder filtrations are operated on these presumptive SNPs which narrowed them down to two single SNPs (data not shown) as these results will be reported separately in another publication aimed at addressing the functionality of those genes. Recently, we also reported and published karyotyping and chromosome HeatMap of some regional genomic of interest from the same CKD patients [17].

Here we sought to establish and describe the relationship between PAM matrices obtained from these CDK patients that we separately established and compared them to 1000 genome and genomAD dataset matrices thus allowing us to derive and compare the combined mutability of each amino acids derived from CDK patients as well as the amino acids from the two datasets, 1kG and genomAD. These measures allow us to establish the ranking mutabilities of the amino acids among the three data sets as well as a profile of amino acids mutations that can be damageable in disease situation like CKD.

Objective

Specific Objectives

This study has four objectives:

1. To sequence the human genome from patients suffering from chronic kidney disease, on

Illumina-MiSeq platform.

2. Exploring the possibilities of correlation between plasma and urinary kidney profile and some biochemical parameters, as well of association between known phenotypic SNPs and amino acids exchange.

3. Investigating the existence of deleterious amino acids exchange.

4. Authenticating these exchanges on phenotypic basis by reconstructing the variant genes harboring these mutations paving the way for future development of animal models for ex vivo studies.

Genomic DNA library preparation and sequencing

• High quality gDNA was purified from lymphocytes whole as described elsewhere. The construction of the library was conducted using the Nextera Rapid Capture kits (Illumina)

• 50 ng of gDNA was utilized using transposase-based chemistry to create exome libraries

•Library was Qc on Bioanalyzer 2100, (Agilent Technology, Santa Clara, CA) to assess the integrity and size library fragments.

• 2 x 75 paired end cycles were performed (+1 cycle to each forward and reverse read to allow for phasing/pre-phasing).

Bioinformatics analyses

The flow chart for bioinformatics analysis have been published by our team elsewhere.

Below are summarized successive steps we used in bioinformatics analysis.

Read Trimming and Clipping of Adapters

Reads are trimmed from the 3' end to have at least a phred score of 30. Illumina sequencing adapters are removed from the reads, and all reads are required to have a length of at least 50 bp. Trimming and clipping are performed using Trimmomatic (Bolger, Lohse, and Usadel 2014) [18]. This step generated: Samples Readset, Raw pair Read, number of Paired Reads obtained from the sequencer, the number of Remaining Paired Reads after the trimming step and the percentage of Surviving Paired Reads and Raw Paired Reads.

Read Alignment to a Reference Genome

The filtered reads are aligned to a reference genome. The genome used in this analysis is Homo_sapiens assembly hg19. Each readset is aligned using BAW [Li and Durbin (2010)] [19] which creates a Binary Alignment Map file (bam). Then we merged all readset BAM files from the same sample into a single global BAM file using Picard.

BWA is a fast light-weighted tool which aligns relatively short sequences (queries) to a sequence database (target), such as the human reference genome. It is based on the Burrows-Wheeler Transform (BWT). BWA is designed for short queries up to ~200 bp with low error rate (< 3%). It does gap global alignment, supports paired-end reads, and is one of the fastest short read alignment algorithms to date while also visiting suboptimal hits.

Read Alignment Improvement

Because of lack of perfect aligner tool, the alignment step remains really sensitive to the aligner parameters as well as technical and biological variations. To increase the quality and speed of subsequent variant calling, we performed a series of procedures to improve the alignment. These procedures consist of realigning the surrounding short insertions or deletions, fixing possible read mate discrepancy due to realignment marking duplicated reads and recalibrating read base quality.

Realigning Short Insertions and Deletions (INDELs)

The realignment is done using GATK (DePristo et al. 2011) [20]. INDELs in reads (especially near the ends) can trick the mappers into mis-aligning with mismatches. These artifacts resulting from mismatches can harm base quality recalibration and variant detection. Realignment around INDELs helps to improve the accuracy of several downstream steps.

Insertion and deletion realignment is performed on regions where multiple base mismatches are preferred over INDELs by the aligner since it appears to be less costly for the algorithm. Such regions will introduce false positive variant calls which may be filtered out by realigning those regions properly. Mainly realignment will occur in 3 different region types: (1) Known sites (e.g., dbSNP, 1000 Genomes), (2) INDELs seen in original alignments (in CIGARs) (3) sites where evidence suggests a hidden INDEL.

Fixing Read Mates

Once local regions are realigned, the read mate coordinates of the aligned reads need to be recalculated since the reads are realigned at positions that differ from their original alignment. The read mate positions fixing is performed using multi-task application Picard and will be subsequently used to mark duplicates.

Marking Duplicates

Because Exome Sequencing experiment encompasses several PCR amplifications steps which could induce coverage bias during the library construction, this bias could generate high level of false positives and false negatives during the variant calling steps (SNV, SV and CNV). Thus, removing or marking read duplicates using Picard is an important step of the sequencing analysis. Aligned reads are duplicates if they have the same 5' alignment positions (for both mates in the case of paired-end reads). All but the best pair (based on alignment score) will be marked as a duplicate in the bam file. Duplicates reads will be excluded in the subsequent analysis.

Base Quality Recalibration

This step allows recalibrating base quality scores of sequencing-by-syntheses reads in an aligned BAM file. After recalibration, the quality scores in the QUAL field in each read in the BAM output are more accurate in that, the reported quality score is closer to its actual probability of mismatching to the reference genome. QUAL represents the accuracy of the genotyping and when QUAL is 20, this means that there is 99% probability that there is a variant at the site. The recalibration process is accomplished by analyzing the covariation among several features of a base. For example: (1) Reported quality score, (2) the position within the read and (3) the preceding and current nucleotide (sequencing chemistry effect) observed by the sequencing machine. The recalibration process is performed using GATK (DePristo et al. 2011) [20]

Sequencing and Alignment Metrics per Sample

The aim of this step is to provide each sample a general summary statistics and Sample readsets are merged for clarity.

SNV Annotations

An in-house database identifies regions in which reads are confidently mapped to the reference genome. Generally, low mappability corresponds to RepeatMasker regions and decreases substantially with the increasing in the read length. A region is identified as HC = coverage too high, LC = low coverage, MQ = too low mean mapQ (< 20) and ND = unmappable region (no data) at the position.

The VCF files are also annotated for:

dbSNP using the SnpSift software (Cingolani, Patel, et al. 2012) [21].
Variant effects (predicted variant effects on gene, such as amino acid changes) using the SnpEff software (Cingolani, Platts, et al. 2012) [21].

The SnpSift http://snpeff.sourceforge.net/SnpSift.html software which is a toolbox used to filter and localize missense which are number of variants in coding region for which the variation generates a codon modification. This lead to generate two types of variant metrics tables: variants for codons changes and amino acids changes induced by SNV.

Variants Metrics

The variant metrics and statistics have been generated using the SnpEff software (Cingolani, Platts, et al. 2012) [21].

Non-synonymous mutations in humans

We query two reference data sets for natural human variants that we used work. The first set of natural variants was from the 1000 Genomes Project (1kG), and the second from gnomAD. The 1kG data comprises a compilation of human DNA variants from 1092 individuals from diverse populations across the world. Only variants with a frequency of occurrence of >1% were included. To avoid the presence of unstable elements, major histocompatibility and immunoglobulin were removed from group of the transcripts. Of interest here were the variants that mapped to protein coding regions as synonymous (sSNPs) and non-synonymous (nsSNPs) variants. These were obtained from Ensembl [22 40] v67, as described in de Beer et al [1].

The other source of natural variants was gnomAD, which contains data from various disease-specific and population genetic studies. We used v.2.0 which had variants from over 140,000 individuals. The data were downloaded from the Broad Institute’s web site. As with the 1kG data, the genome coordinates had to be converted to the corresponding sSNPs and nsSNPs in the relevant protein coding regions to get the amino acid changes involved.

Non-synonymous mutations in humans from CKD Patients

As initially reported in section bioinformatics analysis, each transcript representative for each protein sequence obtained from the VCF files was allocated Ensembl ID using the mapping tool provided by Uniprot. We derived nsSNPs for each protein from CKD from VCF files using SnpEff software and the metrics converted into two tables with one representing codon change and the other one amino acids exchange with each record of entry on Y-axis symbolized by one single letter amino acid. Because, major histocompatibility complex and immunoglobulins are inherently unstable, they were removed from the pool of the transcripts The data set representative of amino acids exchanges is later converted into amino acid pair score matrices or exchange frequencies among amino acids.

Statistical Analysis

We calculated the Pearson correlation coefficient (r_P) to examine the linear relationship between the variables and the Spearman correlation coefficient (r_S) to examine the monotonous relationship over sequence numbers. We interpreted the obtained correlations data as strong relationship if 0.7 < |r| ≤ 1, as moderate relationship when 0.4 < |r| ≤ 0.7, and for weak or absence of relationship if 0 ≤ |r| ≤ 0.4. We analyzed the differences in the mutation numbers between the Patients, 1kg on one hand the patients and gnomAD datasets using Fisher's exact test. We applied Bonferroni correction to the p-values when examining multiple comparisons. We obtained all analysis results and graphics using the R package program (R Core Team, 2021) [23].

Resources: R Core Team (2021). R: A language and environment for statistical computing.

We will have weak correlation, if 0 ≤ |r| ≤ 0.4, moderate correlation, if 0.4 < |r| ≤ 0.7 and strong correlation if 0.7 < |r| ≤ 1

This project was carried out to compare the amino acids exchange from CKD patients databases compared to the amino acids exchanges from 1000 genome and genomAD databases. The data from CKD patients were obtained by carrying out exome sequencing on genomic DNA libraries built from the patients genomic DNA (see methods). We already published the biochemical profiles from these patients in Eur. J. Med. Health. Sci [17}. The Average distribution of capture sub regions from exome sequencing is presented on Fig. 1. This distribution of mutation sites within genomic regions from different variants in CKD patients is distributed to nine genomic regions. This statistical reparation of the variants may represent a transition bias observed in introns, pseudogenes, and other noncoding regions (Gojobori et al. 1983;) [24] The biochemical pipelines and ANOVAR tool we used for the bioinformatic analysis of the FASTQ data are presented were published elsewhere [17]. These were instrumental in experimental determination nsSNPs which were filtered and annotated with respect to their functions. In Fig. 4, we presented the codons HeatMap changes in CKD patients. The codons exchange frequencies in the transcripts reflect the cumulative distribution of codons mutability for silent (blue area), missense (yellow spots) nonsense (red spots) and this distribution of mutability itself depends on the absence or presence CpG dinucleotides islands. The inset colors show the probability density distribution by the type of mutation. The variation in colors reflect the degree of exchanges: blue color signifies low exchanges whereas red color is associated with high exchanges. In Table 1, we presented the amino acids exchanges in CKD patients, together with the amino acids’ exchanges derived from the 1000 genome and genomAD databases.

The amino acids exchanges originating from CKD patients, database, 1000 genome and genomAD databases.

We carried out the exome sequencing project in patients with CKD in part to retrieve all nsSNPs and their associated transcripts. These SNPs were submitted to successive bioinformatic analysis and underwent through many steps of hard filtering (see method) to ascertain that, just all pathological SNPs of interest were retained from which we produced a square matrice 20x20 of amino acids substitution from CKD patients. We also simultaneous carried out analysis of amino acids exchange matrix generated from 1091 individuals obtained from 1000 genome dataset and 140,000 individuals from genomAD database. The CKD matrix combined with the 1 kg matrix and the genomAD matrix presented in Table 1 depicted an asymmetric configuration in term of topology which itself is inherent to the difference in mutability of different codons and the pathological state of different proteins in CKD samples [25, 26]. This difference in mutability is due in part by flanking nucleotides context of the mutation site. Indeed, it has been reported that up one heptamers, 3 nucleotides upstream and 3 nucleotides downstream of the site of nsSNPs, can affect the mutation [27–30]. For example, the elevation in mutation rate is due to methyl cytosine deamination at CpG dinucleotides [31] and more specifically, mutability stemming from nonsynonymous mutations depends on the specificality of the amino acids that are exchanged, reflecting differences in both physicochemical properties [32] and metabolic effects [33].

Mutability of amino acids in CKD, 1000 genome and genomAD Matrix

The mutability of the amino acids in the three databases are summarized on Table 1. Each cell contains information about each amino acid from the three databases, with on the top, the CKD, in the middle the 1 kg and genomAD in the bottom and the subsequent presentation of the rest of our results in this table will follow this order. Talking about the mutability of the amino acids among the three databases, Arginine remains the most mutable amino acid all tree databases with the mutability metrics organized respectively from the top to the bottom in its cell is (0.022, 0.030 and 0.021) for the three databases. Inside the cell, Arginine mutability from 1000 genome dataset remains higher vis a vis to Arginine mutability in CKD database which itself is higher than Arginine mutability in genomAD dataset. However, those amino acids with more chemical complexity structures like: Tryptophan and Phenylalanine respectively depict the lowest mutabilities values for the three datasets (CKD, 1 kg and genomAD) with Tryptophan mutability being (0.007, 0.004 and 0.006) respectively for the three databases and Phenylalanine mutability being (0.005, 0.004 and 0.006) respectively.

We also compared the probabilistic distribution of amino acids in human proteome between the three datasets (see figure S1). On Y axis is shown the number of mutations of amino acids and on X axis we presented the frequency of occurrence of the amino acids in human proteome. From this result, it is clear that Arginine and Leucine represent the outliers in that they deviate from the common alignment on the axe from other amino acids. The result also shows moderate correlation between the mutability and the probabilistic distribution of the amino acids in human proteome in the three datasets with their correlation coefficients being: (r_CKD= 0.6062, r_1kg = 0.4808, r_genomAD= 0.6430). The same moderate positive correlation is also observed in (figure S2). The analysis of data from 1 kg dataset presents the relationship between mutabilities and the number of codons with the correlation coefficient being (r_1kg = 0.4334) whereas data from CKD and genomAD datasets show week correlation with the correlation coefficients representing the relation between their mutabilities and their number of codons being (r_CKD= 0.3108, r_genomAD= 0.2852) respectively. Contrarily to moderate and weak correlation values observed between the number of codons and their mutabilities on one hand, on the other hand, the comparison of mutabilities between the three datasets presented on Fig. 3 shows strong correlation with their respective correlation coefficient comparison being: (r_{CKD vs 1kg} = 0.9225, r_{CKD vs genomAD} = 0.9431 and r_{genomAD vs 1kg} = 0.9486). This is a clear demonstration that, on the evolutionary basis, the distance between proteins from the three datasets is close since it is generally accepted that more substitutions of amnio acids are needed to change one evolutionary sequence into another one and the more substitutions we have, the larger becomes the evolutionary distance [34]. Besides, the evolutionary property also depends on some amino acid properties like isoelectric point, polarity and volume [35].

The close evolution distance between the three databases as observed on Fig. 3 is also reiterated on (figure S3) where we show strong correlation between the mutability and the number of mutations at least in CKD and 1 kg datasets and moderate correlation in genomAD database, with 1000 genome dataset being (r_{1000 genome} = 0.8702), followed by CKD patients (r_CKD = 0.7002) moderate correlation in genomAD dataset (r_genomAD = 0.6442).

The figure S4 compares the number of mutations to the natural occurrence of amino acids in human proteome and the three databases depicted moderate correlation between the number of mutations and the natural occurrence of amino acids. Indeed, it was reported that variability in mutation is context-depend [36–38], and it is well known that, the prediction of mutation variability rate in the human genome is linked to the environment surrounding the sites of mutation [36–38]. Voight and Aggarwala [36] reported from the variant site that, three nucleotides upstream positions and downstream positions surrounding the site of the mutation affect the mutation, which means that up to 7-mer sequence context is detrimental for the mutation to take place.

Nature of amino acids exchange in CKD, 1000 genome and genomAD databases

The profile of one single amino acid changing all other 19 amino acids in human proteome in all three datasets is portrayed on Fig. 4 and this is depicted as % of change one amino acid to all other amino acids and the frequency of the occurrence of different exchanges is clearly depicted between different amino acids. For example, in the CKD patient’s dataset, more mutations are recorded for: Tryptophan (W), Histidine (H), Glycine (G) and Lysine (K) and these four later four amino acids have Arginine as end product in term of mutability (p < 0.0001) and Methionine (M) and Isoleucine (I) both mostly being changed into valine (V) (p < 0.0001). We should recall that (M and I) are hydrophobic amino acids with aliphatic chain. In the concept of exchangeability, (I) which is hydrophobic aliphatic amino acid will preferably be substituted by another hydrophobic amino acid, particularly aliphatic amino acids and this assertion is true for M too [39]. On the structural basis, being hydrophobic, (I) prefers to be buried in hydrophobic protein cores plus being Cβ-branched like (V) and (T). Then, in substrates recognition and hydrophobic ligands binding basis, (I) and M can easily be substituted to V [39].

The amino acid substitutions in diseases have well been documented [40]. Amino acids share common properties and several classifications has been adopted and sometimes, it is difficult to put all amino acids of the same group into a one invariant group because, each amino acid in a different protein environment will have different characteristics and one cannot a priori determine its role in the protein based on hydrophobicity or polarity and no-polarity nature.

On large scale of naturally occurring of human amino acid variations, it was demonstrated that there is a preference for some amino acid substitutions to be associated with diseases as it is frequently observed for the Arginine [40].

The other way of classifying amino acids is the use of mutation substitution matrix. High value in this matrix is indicative that substitution often seen in nature is favorable and vice versa. Mutation matrices are used as a rough guide for how good or bad a particular change will be. Another useful aspect of this concept is that, the matrix can be calculated for different datasets, most especially in diseases situation. We have generated the mutation matrix for CKD and came up with particular specific in amino acids changes with four amino acids: Tryptophan (W), Histidine (H), Glycine (G) and Lysine (K) all together changing to Arginine (R) and this confirmed previous past observation with Arginine frequently associated with diseases [40].

(W) is an aromatic amino acid, but its uniqueness in term of chemistry and size means that its replacement by anything else will be disastrous and this the situation we are encountering in CKD. This drastic situation in term of changeability between amino acids of different groups is seen with (H) too which is generally considered to be a polar amino acid, however its uniqueness with regard to its chemical properties makes of (H) not a particularly good candidate for substitution with any other amino acid, however, (H) was also changed to (R) which is a positively-charged, polar amino acid.

Arginine (R) generally prefers to be on the surface of the protein, but its amphipathic nature means that part of the side chain is buried.

Lysine (K) can be substituted by arginine or other polar amino acids. In the determination of the involvement of (K) biological activity in protein function, the substitution of (K) with (R) completely eliminates the biological activity of TGF-α [41].

Amino acid mutability Hierarchy Scores

The mutability scores of the amino acids are presented on Fig. 5 and among the three databases, the most mutable amino acids remain again the arginine (R). Arginine (R) besides of being encoded by six codons, it harbors two CpG units and has thus has higher the potential propensity mutability and targetability as compared to (M) which is coded by one single codon, but still performs well in term of mutability score.

Distribution of average protein mutabilities for all human proteins in the 1 kG, genomAD and CKD patients

The Fig. 6 presents the average distribution of proteins mutability shows that all amino acids proportionally show higher mutability in genomAD dataset compared to 1 kg and CKD datasets and the average proteins mutability in CKD as well as in 1 kg are equally comparable. This higher mutability in genomAD is inherent to the gene fragments contained in this dataset as it harbors genomic variants as well as transcript variants.

Chronic kidney disease (CKD) is defined as abnormalities in kidney structure or function and requires either GFR below 60 ml/min/1.73 m² or markers of kidney damage, including albuminuria [42]. It is a congruence of many vascular diseases with patients having dysfunctions of many physiological pathways. The pathology is infiltrated by other diseases like: diabetes, hypertension [43–45], cardiovascular morbidity [46, 47] as well as many genetic abnormalities like genomic instability [47, 48–51]and instability which is later translated into chromosomal damage and ultimately culminating to renal pathologies. Recent progress in genome-wide association studies revealed SNPs linked to different pathologies associated to CKD, like hypertension [52], coronary artery disease [53], subclinical vascular disease [54] and overlapping between genetic variants observed in CKD patients and cardiovascular pathologies [55]. The goal of our study is to establish a repertoire of allelic variants and susceptibility to kidney diseases and ultimately find some candidate genes directly linked to CKD. To this end, we run exome sequencing from CKD patient samples and determined the average mutation distribution sites within genomic regions from different variants. After bioinformatic analysis coupled with filtration and hard filtration, we came up with two single genes that are common to all patients (data not shown as we will be reporting this separately). We also derived the substitutions or exchanges between the amino acids, their mutabilities as well as the probabilistic distribution of these amino acid exchanges in human proteome. Our results in this study provides a unique relation between amino acids exchanges status in CKD. The results as presented on Table 1 clearly show a statistically difference between some individual amino acids exchange matrix in CKD patients as compared with the natural variants, we derived from 1000 genome and genomAD datasets. Our data corroborate with those reported by De Beer et al [1] that the amino acids change in disease-associated variants differed radically from variants obtained from 1000 genome dataset. In this study, we went one step forward to establish a net difference in variation between amino acids exchanges deriving from CKD patients to those obtained from 1000 genome and genomAD datasets. This is a first demonstration as such in the scope of CKD patients using a square PAM matrix format 20 X 20 PAM to derive amino acids substitution and their plausible mutabilities in renal disease. In the concept of mutability, it is also known that cytosines in a cytosine–guanine (CpG) dinucleotide context is known for their propensity to mutate at a rate that can be several hundred-fold greater than transversions at other bases, and in genetic variation, these representations are disproportionately represented. It was shown that, hotspots regions of CpG dinucleotides islands play a role in mutational mechanism leading to determination of the amino acids that are prone to mutation. Our data in genomAD in term of mutability presents the same profile as well as the one from 1000 genome dataset as compared with CKD dataset where some cells present statistically significant difference in term of exchangeability between the CKD patients’ dataset. However, the genomAD database shows a proportionally higher mutability for all amino acids compared to 1 kg and CKD datasets and the average proteins mutability in CKD as well as in 1 kg are equally comparable (Fig. 6). This difference is inherent to the nature of the three datasets. First, we obtained the CKD dataset from Exome sequencing experiment on Illumina Platform. Then, the DNA sequencing we are dealing with stemmed from the transcripts although some primers from Illumina kits can sometime overlap with some intervening sequences and thereby introducing possibilities that some genomic sequence may be carried on in the results of sequencing data. Second, whereas the1000 genome database was developed using a combination of multiple sequencing approaches, including: low coverage whole genome sequencing, with depth mean of 7.4X, deep Exome sequencing with depth mean of 65.7X and dense microarray genotyping and the result in this database are believed to emanate from the transcript pools uniquely. Therefore, scientists use 1000 genome data base as reference to run variants for Exome sequencing data. Lastly, the genomAD database in contrary was developed by harmonizing both exome and genome sequencing data from a wide variety of large-scale sequencing projects, then it is a direct combination of Exome sequencing and whole genome sequencing data, and this may explain the higher proportion of mutability index observed in this database comparatively 1000 genome and CKD datasets. Moreover, on genomic scale, the human genome is known to contain ∼30,000 CpG islands, long stretches (CpG repeats) of DNA with unusually elevated levels of CpG dinucleotides with many occurring at genes promoters. CpG islands are also residents of intervening sequence making the genes prone to mutation thus explaining their contribution to diseases This also explains why the distribution of average mutability genomAD as compared with 1 kg and CKD Fig. 6. In our experiment in (figure S1), we compared the number of mutations and the probabilistic distribution of amino acids in human proteome. We observed that arginine was the outlier with higher mutability index compared to the rest of amino acids. In fine, after the observation of the natural variants in genomAD as well as 1000 genome databases, it is conclusive to say that the amino acids mutability is mostly driven by the nature and state of the DNA and the mutational mechanisms are favored at the codons harboring the CpG dinucleotides and this explain why Arginine mutates more than twice than other amino acids. This increase in Arginine variants we observed in the three datasets can be explained by the deamination of methylated cytosine (5mC) that results in thymine, which is not readily detected by DNA repair systems. Interestingly, arginine is the only amino acid to contain CpG dinucleotides at the first and second codon positions, which are less redundant than the third position and therefore more likely to lead to an amino acid change. Moreover, codon usage for arginine can have an impact on GC content for a given gene as arginine is encoded by six codons in vertebrate genomes four CGN codons (CGA, CGC, CGG, CGU) and two AGR codons (AGG, AGA). Besides, data from thermodynamic experiments have reported that disease-causing variants tend to destabilize protein structures and their interactions (Khan and Vihinen, 2010; Thusberg, et al., 2011) [56, 57]. Later on, mutagenesis studies have also reported that replacing nonpolar group by polar ones can increase protein solubility. For example, on the study on RNase Sa, it was found that, the most buried carboxylic groups are Asp33 (99% buried) and Asp79 (85% buried) and their respective mutants D33A mutants and D79A are 6kcal/mol and 3.3 kcal/mol less stable than the wild type [58, 59] and replacing Thr 76 by Asp acid increase the protein solubility [59]. In this study, we noticed that four major amino acids, viz: Tryptophan, Histidine, Glycine and Lysine have Arginine in common as a common final product of mutability. Whether this phenomenon is specifically peculiar to kidney diseases, this remains to be investigated thoroughly via mutagenesis studies because each amino acid has its profile of physicochemical characteristic property and can therefore influence the final protein chemical structure.

It is well documented that the determination of protein distances between many sequences necessitates the knowledge of the amino acids substitution rates [26] and the evolutionary pressure also depends on some amino acid physicochemical properties like isoelectric point, polarity and its volume [27]. We analyzed and compared the amino acids substitution rates between our three datasets and observed a strong positive correlation between the mutabilities as observed on Fig. 3 and (figure S3). This positive correlation demonstrates a close evolutionary distance between the proteins from the three datasets since it is generally accepted that more substitutions of amnio acids are needed to change one evolutionary sequence into another one and the more substitutions we have, the larger becomes the evolutionary distance [26 27].

We run Exome sequencing and established a PAM matrix of amino acids changes in CKD patients and compared this matrix to matrices obtained from 1000 genome and genomAD databases. The data demonstrated that the amino acids mutability index in CKD data was statistically significant different as compared with 1000 genome and genomAD datasets. There is strong positive correlation in mutabilities of amino acids distribution in human proteome and the average distribution of mutability is higher in genomAD dataset as compared CKD and 1000 genome. In the CKD dataset, Arginine remains a common product of mutability from four amino acids like: Tryptophan, Histidine, Glycine and Lysine and Arginine was shown to be also the outlier and the driver of mutability in the three data sets. More experiments remained to be conducted, more especially using other diseases context dataset to ascertain the peculiar mutability pattern of arginine. Nonetheless, this study remains the first finding to layout clearly a profile of amino acids changes in CKD as well as a jump in mutability in genomAD dataset.

de Beer, T. A. P. et al. Amino Acid Changes in Disease-Associated Variants Differ Radically from Variants Observed in the 1000 Genomes Project Dataset. PLoS Comput. Biol. 9, (2013).
Lohmueller, K. E., Mauney, M. M., Reich, D. & Braverman, J. M. Variants associated with common disease are not unusually differentiated in frequency across populations. Am. J. Hum. Genet. 78, 130–136 (2006).
Witte, J. S., Visscher, P. M. & Wray, N. R. The contribution of genetic variants to disease depends on the ruler. Nat. Rev. Genet. 15, 765–776 (2014).
Zheng-Bradley, X. & Flicek, P. Applications of the 1000 Genomes Project resources. Brief. Funct. Genomics 16, 163–170 (2017).
Jablonka, E. The evolutionary implications of epigenetic inheritance. Interface Focus 7, (2017).
Karczewski, K. J. et al. The mutational constraint spectrum quantified from variation in 141,456 humans. Nature 581, 434–443 (2020).
Minikel, E. V. et al. Evaluating drug targets through human loss-of-function genetic variation. Nature 581, 459–464 (2020).
Collins, R. L. et al. A structural variation reference for medical and population genetics. Nature 581, 444–451 (2020).
Mallett, A. et al. The prevalence and epidemiology of genetic renal disease amongst adults with chronic kidney disease in Australia. Orphanet J. Rare Dis. 9, 1–9 (2014).
Freedman, B. I., Soucie, J. M. & Mcclellan, W. M. Family history of end-stage renal disease among incident dialysis patients. J. Am. Soc. Nephrol. 8, 1942–1945 (1997).
Connaughton, D. M. et al. The Irish Kidney Gene Project - Prevalence of Family History in Patients with Kidney Disease in Ireland. Nephron 130, 293–301 (2015).
Genovese, G. et al. Association of trypanolytic ApoL1 variants with kidney disease in African Americans. Science (80-. ). 329, 841–845 (2010).
Gast, C. et al. Collagen (COL4A) mutations are the most frequent mutations underlying adult focal segmental glomerulosclerosis. Nephrol. Dial. Transplant. 31, 961–970 (2016).
Deutsch, A.J., Ahlqvist, E. & Udler, M.S. Phenotypic and genetic classification of diabetes. Diabetologia 65, 1758–1769 (2022). https://doi.org/10.1007/s00125-022-05769-4
Cameron-Christie, S. et al. Exome-based rare-variant analyses in CKD. J. Am. Soc. Nephrol. 30, 1109–1122 (2019).
Lata, S. et al. Whole-Exome Sequencing in Adults With Chronic Kidney Disease: A Pilot Study. Ann. Intern. Med. 168, 100–109 (2018).
Fazaludeen, M. F., Warille, A. A., Alaraj, M. I. & Nuglozeh, E. Chromosome HeatMap in CDK Patients as Defined by Multiregional Sequencing on Illumina MiSeq Platform. Eur. J. Med. Heal.
Bolger, A. M., Lohse, M. & Usadel, B. Trimmomatic: A flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120 (2014).
Li, H. & Durbin, R. Fast and accurate long-read alignment with Burrows-Wheeler transform. Bioinformatics 26, 589–595 (2010).
DePristo, M. A. et al. A framework for variation discovery and genotyping using next-generation DNA sequencing data. Nat. Genet. 43, 491–498 (2011).
Cingolani, P. et al. Using Drosophila melanogaster as a model for genotoxic chemical mutational studies with a new program, SnpSift. Front. Genet. 3, 1–9 (2012).
Howe, K. L. et al. Ensembl 2021. Nucleic Acids Res. 49, D884–D891 (2021).
Bryant Jeri L, Boughter John D, Gong Suzhen, LeDoux Mark S, Heck Detlef H et al. HHS Public
Gojobori T. Codon Substitution in Evolution and the Saturation of Synonymous Changes. Genetics. 1983; 105: 1011-1027
Jordan, I. K. et al. Protein evolution: Causes of trends in amino-acid gain and loss (Reply). Nature 442, 9–10 (2006).
Jordan, I. K. et al. A universal trend of amino acid gain and loss in protein evolution. Nature 433, 633–638 (2005).
Sved, J. & Bird, A. The expected equilibrium of the CpG dinucleotide in vertebrate genomes under a mutation model (DNA methylation/factor IX/a-globin gene/pseudogene/neutral hypothesis). Evolution (N. Y). 87, 4692–4696 (1990).
Blake, R. D., Hess, S. T. & Nicholson-Tuell, J. The influence of nearest neighbors on the rate and pattern of spontaneous point mutations. J. Mol. Evol. 34, 189–200 (1992).
Hess, S. T., Blake, J. D. & Blake, R. D. Wide variations in neighbor-dependent substitution rates. J. Mol. Biol. 236, 1022–1033 (1994).
Morton, B. R., Oberholzer, V. M. & Clegg, M. T. The influence of specific neighboring bases on substitution bias in noncoding regions of the plant chloroplast genome. J. Mol. Evol. 45, 227–231 (1997). Science (80-. ). 185, 862–864 (1974).
Akashi, H. & Gojobori, T. Metabolic efficiency and amino acid composition in the proteomes of
Nowicka, A. et al. Correlation between mutation pressure, selection pressure, and occurrence of amino acids. Lect. Notes Comput. Sci. (including Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinformatics) 2658, 650–657 (2003).
Li, W.-H. & Xia, X. What Amino Acid Properties Affect Protein Evolution? J. Mol. Evol. 47, 557–564 (1998).
Aggarwala, V. & Voight, B. F. An expanded sequence context model broadly explains variability in polymorphism levels across the human genome. Nat. Genet. 48, 349–355 (2016).
Rahbari, R. et al. Timing, rates and spectra of human germline mutation. Nat. Genet. 48, 126–133 (2016).
Michaelson, J. J. et al. Whole-genome sequencing in autism identifies hot spots for de novo germline mutation. Cell 151, 1431–1442 (2012).
Betts, M. J. & Russell, R. B. Amino Acid Properties and Consequences of Substitutions. in Bioinformatics for Geneticists 289–316 (John Wiley & Sons, Ltd, 2003). doi:https://doi.org/10.1002/0470867302.ch14.
Petukh, M., Kucukkal, T G. Alexov, E. On human disease-causing amino acid variants: statistical study of sequence and structural patterns. Hum. Mutat. 36, 524–534 (2015).
Defeo-Jones, D. et al. Substitution of lysine for arginine at position 42 of human transforming growth factor-alpha eliminates biological activity without changing internal disulfide bonds. Mol. Cell. Biol. 9, 4083–4086 (1989).
Gojobori, T., Li, W. H. & Graur, D. Patterns of nucleotide substitution in pseudogenes and functional genes. J. Mol. Evol. 18, 360–369 (1982).
Wickham, H. et al. Welcome to the Tidyverse. J. Open Source Softw. 4, 1686 (2019).
Smyth, L. J., Duffy, S., Maxwell, A. P. & McKnight, A. J. Genetic and epigenetic factors influencing chronic kidney disease. Am. J. Physiol. - Ren. Physiol. 307, 757–776 (2014).
Tonelli, M., Sacks, F., Pfeffer, M., Jhangri, G. S. & Curhan, G. Biomarkers of inflammation and progression of chronic kidney disease. Kidney Int. 68, 237–245 (2005).
Snively, C. S. & Gutierrez, C. Chronic kidney disease: Prevention and treatment of common complications. Am. Fam. Physician 70, 1921–1930 (2004).
Adler, A. I. et al. Development and progression of nephropathy in type 2 diabetes: The United Kingdom Prospective Diabetes Study (UKPDS 64). Kidney Int. 63, 225–232 (2003).
El Nahas, M. et al. The global challenge of chronic kidney disease. Kidney Int. 68, 2918–2929 (2005).
Levey, A. S. et al. Chronic kidney disease as a global public health problem: Approaches and initiatives - A position statement from Kidney Disease Improving Global Outcomes. Kidney Int. 72, 247–259 (2007).
Sandoval, S. B. et al. Genetic damage in chronic renal failure patients is associated with the glomerular filtration rate index. Mutagenesis 25, 603–608 (2010).
Sandoval, S. B. et al. Genomic instability in chronic renal failure patients. Environ. Mol. Mutagen. 53, 343–349 (2012).
Lialiaris, T. et al. Chromosome instability in patients with chronic renal failure. Genet. Test. Mol. Biomarkers 14, 37–41 (2010).
Moffitt, T., Hariton, F., Devlin, M., Garrett, P. J. & Hannon-Fletcher, M. P. A. Oxidative DNA Damage Is Elevated in Renal Patients Undergoing Haemodialysis. Open J. Prev. Med. 04, 421–429 (2014).
Ehret, G. B. et al. Genetic variants in novel pathways influence blood pressure and cardiovascular disease risk. Nature 478, 103–109 (2011).
Schunkert, H. et al. Large-scale association analysis identifies 13 new susceptibility loci for coronary artery disease. Nat. Genet. 43, 333–340 (2011).
Bis, J. C. et al. Meta-analysis of genome-wide association studies from the CHARGE consortium identifies common variants associated with carotid intima media thickness and plaque. Nat. Genet. 43, 940–947 (2011).
Olden, M. et al. Overlap between common genetic polymorphisms underpinning kidney traits and cardiovascular disease phenotypes: The CKDGen consortium. Am. J. Kidney Dis. 61, 889–898 (2013).
Khan, S. & Vihinen, M. Performance of protein stability predictors. Hum. Mutat. 31, 675–684 (2010).
Thusberg, J., Olatubosun, A. & Vihinen, M. Performance of mutation pathogenicity prediction methods on missense variants. Hum. Mutat. 32, 358–368 (2011).
Thurlkill, R. L., Grimsley, G. R., Scholtz, J. M. & Pace, C. N. Hydrogen Bonding Markedly Reduces the pK of Buried Carboxyl Groups in Proteins. J. Mol. Biol. 362, 594–604 (2006).
Trevino, S. R. et al. Asp79 makes a large, unfavorable contribution to the stability of RNase Sa. J. Mol. Biol. 354, 967–978 (2005).

Table 1. Observation of amino acids exchanges in human proteins variants. The data representing the CKD patients are on the top in the row of each cell, 1kG in the middle and genomAD in the bottom. Amino acids are arranged according to one single letter nomenclature following the scale of Fauchère and Pliska according to their increasing hydrophobicity from left to right and the top to the bottom. The less hydrophobic are on the left and the most on the right. The yellow blocks indicate absence of a statistically significant difference between the CKD patients and the two data sets, the light blue cells mean there is a statistically significant difference between the CKD patients data set and 1kG, genomAD or 1 kG and genomAD together. In the column representing the total number of mutations, the green color denotes absence of mutation between the CKD patients and the two data sets while the light brown color indicates presence of significant difference between the CKD patients and the two data sets.

from	To																				Total	MUT
from	R	K	D	E	N	Q	S	G	H	T	A	P	Y	V	M	C	L	F	I	W	Total	MUT
R		709	0	1	2	2349	476	773	2071	137	3	280	0	0	80	1374	371	0	63	1048	9737	0.022
		738	0	0	0	4630	553	761	4125	247	0	216	0	0	71	3420	467	0	137	2695	18060	0.030
		29213	0	0	0	118098	31134	51256	90111	13721	0	23584	0	0	2910	89234	33386	0	4590	73483	560720	0.021
K	907		0	740	392	238	9	0	0	199	8	4	0	0	51	0	3	2	52	0	2605	0.006
	1151		0	843	689	346	0	0	0	300	0	0	0	0	87	0	0	0	84	0	3500	0.006
	54592		0	44799	37900	18060	0	0	0	18093	0	0	0	0	5339	0	0	0	4017	0	182800	0.007
D	1	0		793	1006	0	0	551	242	1	181	0	203	162	0	0	2	0	0	0	3142	0.008
	0	0		855	1994	0	0	780	419	0	153	0	378	379	0	0	0	0	0	0	4958	0.010
	0	0		45423	75306	0	0	43503	25077	0	10473	0	21961	21126	0	0	0	0	0	0	242869	0.011
E	2	1219	890		3	485	0	506	0	0	276	0	0	175	0	0	1	0	0	0	3557	0.006
	0	2682	983		0	631	0	628	0	0	319	0	0	241	0	0	0	0	0	0	5484	0.007
	0	108957	51791		0	39030	0	39604	0	0	19204	0	0	14615	0	0	0	0	0	0	273201	0.008
N	3	482	560	0		0	1336	1	159	182	0	0	94	0	0	0	0	0	105	0	2922	0.010
	0	546	516	0		0	2030	0	236	170	0	0	133	0	0	0	0	0	158	0	3789	0.010
	0	32632	26842	0		0	71753	0	12805	10421	0	0	7260	0	0	0	0	0	9553	0	171266	0.010
Q	1262	314	0	354	0		1	0	724	2	0	261	0	0	0	0	140	0	1	0	3059	0.008
	875	351	0	521	0		0	0	764	0	0	243	0	0	0	0	192	0	0	0	2946	0.006
	49660	16788	0	27915	0		0	0	39814	0	0	16273	0	0	0	0	10064	0	0	0	160514	0.007
S	455	4	4	1	1061	0		781	0	644	468	777	161	2	0	374	669	420	167	49	6037	0.009
	648	0	0	0	1103	0		659	0	637	273	541	284	0	0	673	1386	816	242	44	7306	0.008
	39566	0	0	0	47105	0		35001	0	33559	15717	30612	13424	0	0	34040	42111	42633	12231	4466	350465	0.009
G	1229	0	633	644	0	0	1226		0	0	495	0	0	451	0	148	1	0	0	55	4882	0.009
	1957	0	852	804	0	0	1598		0	0	469	0	0	630	0	197	0	0	0	84	6591	0.009
	89095	0	42935	39897	0	0	61049		0	0	31884	0	0	34963	0	13229	0	0	0	6421	319473	0.010
H	925	0	183	0	124	429	0	0		0	2	122	497	0	0	0	91	0	0	0	2373	0.011
	863	0	130	0	166	486	0	0		0	0	145	869	0	0	0	188	0	0	0	2847	0.010
	44061	0	8282	0	8587	25300	0	0		0	0	10406	37655	0	0	0	9487	0	0	0	143778	0.012
T	208	182	3	0	271	1	667	1	3		1620	312	2	0	1197	0	3	0	1066	0	5536	0.013
	206	200	0	0	389	0	740	0	0		1420	186	0	0	2272	0	0	0	1781	0	7194	0.013
	12803	11219	0	0	19200	0	38021	0	0		64546	15067	0	0	41751	0	0	0	76306	0	278913	0.011
A	2	0	288	250	0	9	775	517	0	2961		473	0	2162	0	1	2	0	0	0	7440	0.013
	0	0	309	205	0	0	892	509	0	3529		357	0	3224	0	0	0	0	0	0	9025	0.012
	0	0	17372	12308	0	0	43999	33995	0	122085		26570	0	113525	0	0	0	0	0	0	369854	0.011
P	441	0	0	0	0	174	1350	1	253	521	539		0	0	0	0	1879	0	0	0	5158	0.011
	502	0	0	0	0	198	2086	0	261	650	666		0	0	0	0	3309	0	0	0	7672	0.012
	40224	0	0	0	0	10781	97490	0	14115	35689	41654		0	0	0	0	117791	0	0	0	357744	0.012
Y	0	0	86	0	55	0	103	0	369	0	0	0		2	0	506	4	132	0	0	1257	0.006
	0	0	62	0	81	0	105	0	490	0	0	0		0	0	1125	0	198	0	0	2061	0.007
	0	0	4627	0	5331	0	6833	0	24730	0	0	0		0	0	51258	0	9737	0	0	102516	0.008
V	0	0	95	93	0	4	2	194	0	6	1193	1	0		1383	1	791	169	2436	0	6368	0.014
	0	0	95	98	0	0	0	191	0	0	932	0	0		2319	0	989	286	3057	0	7967	0.013
	0	0	5746	6574	0	0	0	16368	0	0	49590	0	0		71113	0	59831	16149	79863	0	305234	0.011
M	55	76	0	0	3	0	2	0	1	867	0	0	0	1024		0	293	0	576	0	2897	0.017
	104	107	0	0	0	0	0	0	0	705	0	0	0	1050		0	335	0	826	0	3127	0.014
	7773	7449	0	0	0	0	0	0	0	38736	0	0	0	50830		0	20050	0	40683	0	165521	0.016
C	527	0	0	0	0	0	282	131	0	0	2	0	370	0	1		5	120	1	89	1528	0.008
	350	0	0	0	0	0	260	116	0	0	0	0	594	0	0		0	169	0	82	1571	0.006
	20075	0	0	0	0	0	14644	8198	0	0	0	0	27023	0	0		0	9466	0	5973	85379	0.008
L	244	0	0	1	0	115	303	0	112	4	3	977	1	932	258	0		878	237	58	4123	0.005
	227	0	0	0	0	145	279	0	83	0	0	692	0	1078	304	0		1366	387	95	4656	0.004
	20625	0	0	0	0	8890	13829	0	6199	0	0	49877	0	63501	14570	0		66499	15386	4563	263939	0.006
F	0	0	2	0	8	0	324	0	0	0	0	0	128	114	0	128	728		126	0	1558	0.005
	0	0	0	0	0	0	291	0	0	0	0	0	124	159	0	144	880		117	0	1715	0.004
	0	0	0	0	0	0	17184	0	0	0	0	0	6637	10060	0	9861	50019		6564	0	100325	0.006
I	25	36	0	0	107	0	100	0	1	987	0	1	0	1994	451	0	244	96		0	4042	0.012
	35	30	0	0	180	0	131	0	0	1468	0	0	0	2189	690	0	337	242		0	5302	0.011
	1982	1975	0	0	10201	0	7606	0	0	59179	0	0	0	91395	32522	0	18329	14037		0	237226	0.012
W	438	0	0	0	0	0	51	35	0	0	0	0	2	0	0	76	44	0	0		646	0.007
	225	0	0	0	0	0	38	51	0	0	0	0	0	0	0	169	57	0	0		540	0.004
	12369	0	0	0	0	0	3404	3412	0	0	0	0	0	0	0	10097	2928	0	0		32210	0.006
Patients	6724	3022	2744	2877	3032	3804	7007	3491	3935	6511	4790	3208	1458	7018	3421	2608	5271	1817	4830	1299	78867
1kg	7143	4654	2947	3326	4602	6436	9003	3695	6378	7706	4232	2380	2382	8950	5743	5728	8140	3077	6789	3000	106311
gnomAD	392825	208233	157595	176916	203630	220159	406946	231337	212851	331483	233068	172389	113960	400015	168205	207719	363996	158521	249193	94906	4703947

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Amino acids exchange in patients with CKD as obtained via exome sequencing differ radically from those of natural variants 1000 genome and genomAD databases

Status:

Version 1

Abstract

Figures

Background

Methods

Results

Mutability of amino acids in CKD, 1000 genome and genomAD Matrix

Nature of amino acids exchange in CKD, 1000 genome and genomAD databases

Amino acid mutability Hierarchy Scores

Discussions

Conclusions

References

Table

Additional Declarations

Supplementary Files

Status:

Version 1