1 The field of Genome annotation pay a lot of attentions where the ability to collect and analysis genomical data can provide strong indicator for the study of life\cite{Eisen2007}. A lot of genome annotation centres present various types of annotation tools (i.e cost-effective sequencing methods\cite{Bakke2009}) on different annotation levels. Methods of gene finding in annotated genome can be categorized as: Alignment-based, composition based or combination of both\cite{parra2007cegma}. The Alignment-based method is used when we try to predict a coding gene (i.e. Genes that produce proteins) by aligning DNA sequence of gene to the protein of cDNA sequence of homolog\cite{parra2007cegma}. This approache also used in GeneWise\cite{birney2004genewise} with known splicing signals. Composition-based mothod (known as \textit{ab initio} is based on a probabilistic model of gene structure to find genes and/or new genes accoding to the probility gene value, this method like GeneID\cite{parra2000geneid}. In this section, we will consider a new method of finding core genes from large amount of chloroplast genomes, as a solution of the previous method where stated in section two. This method is based on extracting gene features. The question now is how can we have good annotation genome? To answer this question, we need to focusing on studying the annotation's accuracy (systematically\cite{Bakke2009}) of the genome. The general overview of the system is illustrated in Figure \ref{Fig1}.\\
4 \caption{A general overview of the system}
6 \includegraphics[width=0.5\textwidth]{generalView}
10 In Figure 1, we illustrate the general overview of the system. In this system, there are three main stages: \textit{Database, Gene extraction ,} and \textit{relationships}. There are many international nucleotide sequence databases like (GenBank/NCBI in USA at (http://www.ncbi.nlm.nih.gov/genbank/),\\ EMBL-Bank/ENA/EBI in Europe at (http://www.ebi.ac.uk/ena/), and DDBJ in Japon at (http://www.ddbj.nig.ac.jp/)). In our work, the database must be any confident data source that store annotated or unannotated chloroplast genomes. We will consider GenBank/NCBI database as our nucleotide sequences database. Extract Gene Features, we refer to our main process of extracting needed information to find core genome from well large annotation genomes. Thanks to good annotation tool that lead us to extract good gene features. Here, Gene features can be anything like (genes names, gene sequences, protein sequence,...etc). To verify the results from our system, we need to organize and represent our results in the form of (tables, phylogenetic trees, graphs,...,etc), and compare these results with another annotation tool like Dogma\cite{RDogma}. All this work is to see the relationship among our large population of chloroplast genomes and find the core genome for root ancestral node. Furthermore, in this part we can visualize the evolution relationships of different chloroplast organisms.\\
11 The output from each stage in our system will be considered to be an input to the second stage and so on. The rest of this section, in section 3.1, we will introduce some annotation problem with NCBI chloroplast genomes and we will discuss our method for how can we extract useful data. Section 3.2 we will present here our system for calculating evolutionary core genome based on another annotation tool than NCBI.
13 \subsection{Genomes Samples}
14 In this research, we retrieved 107 genomes of Chloroplasts from NCBI where 9 genomes considered as not good. These 99 genomes lies in the 11 types of chloroplast families, divided as 11 for Algues Brunes, 3 Algue Rouges, 17 Algues Vertes, 45 Angiospermes, 3 Brypoytes, 2 Dinoflagelles, 2 Euglenes, 5 Filicophytes, 7 Gymnosperms, 2 Lycophytes, and 1 Haptophytes, as show in Table \ref{Tab1}.\pagebreak
17 \setlength\LTleft{-30pt}
18 \setlength\LTright{-30pt}
19 \begin{longtable}{@{\extracolsep{\fill}}llllllllll@{}}
21 \caption[NCBI Genomes Families]{List of family groups of Chloroplast Genomes from NCBI\label{Tab1}}\\
24 {\textbf{Category}} & {\textbf{Counts}} & {\textbf{Accession No}} & {\textbf{Scientific Name}} \\
27 & & NC\_001713.1 & Odontella sinensis \\
28 & & NC\_008588.1 & Phaeodactylum tricornutum \\
29 & & NC\_010772.1 & Heterosigma akashiwo \\
30 & & NC\_011600.1 & Vaucheria litorea \\
31 & & NC\_012903.1 & Aureoumbra lagunensis \\
32 Algues Brunes & 11 & NC\_014808.1 & Thalassiosira oceanica \\
33 & & NC\_015403.1 & Fistulifera sp \\
34 & & NC\_016731.1 & Synedra acus \\
35 & & NC\_016735.1 & Fucus vesiculosus \\
36 & & NC\_018523.1 & Saccharina japonica \\
37 & & NC\_020014.1 & Nannochloropsis gadtina \\ [1ex]
38 %Entering second group
39 & & NC\_000925.1 & Porphyra purpurea \\
40 Algues Rouges & 3 & NC\_001840.1 & Cyanidium caldarium \\
41 & & NC\_006137.1 & Gracilaria tenuistipitata \\ [1ex]
43 & & NC\_000927.1 & Nephroselmis olivacea \\
44 & & NC\_002186.1 & Mesotigma viride \\
45 & & NC\_005353.1 & Chlamydomonas reinhardtii \\
46 & & NC\_008097.1 & Chara vulgaris \\
47 & & NC\_008099.1 & Oltmannsiellopsis viridis \\
48 & & NC\_008114.1 & Pseudoclonium akinetum \\
49 & & NC\_008289.1 & Ostreococcus tauri \\
50 & & NC\_008372.1 & Stigeoclonium helveticum \\
51 Algues Vertes & 17 & NC\_008822.1 & Chlorokybus atmophyticus \\
52 & & NC\_011031.1 & Oedogonium cardiacum \\
53 & & NC\_012097.1 & Pycnococcus provaseolii \\
54 & & NC\_012099.1 & Pyramimonas parkeae \\
55 & & NC\_012568.1 & Micromonas pusilla \\
56 & & NC\_014346.1 & Floydiella terrestris \\
57 & & NC\_015645.1 & Schizomeris leibleinii \\
58 & & NC\_016732.1 & Dunaliella salina \\
59 & & NC\_016733.1 & Pedinomonas minor \\ [1ex]
60 %Entering fourth group
61 & & NC\_001319.1 & Marchantia polymorpha \\
62 Bryophytes & 3 & NC\_004543.1 & Anthoceros formosae \\
63 & & NC\_005087.1 & Physcomitrella patens \\ [1ex]
65 & & NC\_014267.1 & Kryptoperidinium foliaceum \\
67 & NC\_014287.1 & Durinskia baltica \\ [1ex]
69 & & NC\_001603.2 & Euglena gracilis \\
70 Euglenes & 2 & NC\_020018.1 & Monomorphina aenigmatica \\ [1ex]
71 %Entering seventh group
72 & & NC\_003386.1 & Psilotum nudum \\
73 & & NC\_008829.1 & Angiopteris evecta \\
74 Filicophytes & 5 & NC\_014348.1 & Pteridium aquilinum \\
75 & & NC\_014699.1 & Equisetum arvense \\
76 & & NC\_017006.1 & Mankyua chejuensis \\ [1ex]
77 % Entering eighth group
78 & & NC\_001568.1 & Epifagus virginiana \\
79 & & NC\_001666.2 & Zea Mays \\
80 & & NC\_005086.1 & Amborella trichopoda \\
81 & & NC\_006050.1 & Nymphaea alba \\
82 & & NC\_006290.1 & Panax ginseng \\
83 & & NC\_007578.1 & Lactuca sativa \\
84 & & NC\_007957.1 & vitis vinifera \\
85 & & NC\_007977.1 & Helianthus annuus \\
86 & & NC\_008325.1 & Daucus carota \\
87 & & NC\_008336.1 & Nandina domestica \\
88 & & NC\_008359.1 & Morus indica \\
89 & & NC\_008407.1 & Jasminum nudiflorum \\
90 & & NC\_008456.1 & Drimys granadensis \\
91 & & NC\_008457.1 & Piper cenocladum \\
92 & & NC\_009601.1 & Dioscorea elephantipes \\
93 & & NC\_009765.1 & Cuscuta gronovii \\
94 & & NC\_009808.1 & Ipomea purpurea \\
95 Angiospermes & 45 & NC\_010361.1 & Oenothera biennis \\
96 & & NC\_010433.1 & Manihot esculenta \\
97 & & NC\_010442.1 & Trachelium caeruleum \\
98 & & NC\_013707.2 & Olea europea \\
99 & & NC\_013823.1 & Typha latifolia \\
100 & & NC\_014570.1 & Eucalyptus \\
101 & & NC\_014674.1 & Castanea mollissima \\
102 & & NC\_014676.2 & Theobroma cacao \\
103 & & NC\_015830.1 & Bambusa emeiensis \\
104 & & NC\_015899.1 & Wolffia australiana \\
105 & & NC\_016433.2 & Sesamum indicum \\
106 & & NC\_016468.1 & Boea hygrometrica \\
107 & & NC\_016670.1 & Gossypium darwinii \\
108 & & NC\_016727.1 & Silene vulgaris \\
109 & & NC\_016734.1 & Brassica napus \\
110 & & NC\_016736.1 & Ricinus communis \\
111 & & NC\_016753.1 & Colocasia esculenta \\
112 & & NC\_017609.1 & Phalaenopsis equestris \\
113 & & NC\_018357.1 & Magnolia denudata \\
114 & & NC\_019601.1 & Fragaria chiloensis \\
115 & & NC\_008796.1 & Ranunculus macranthus \\
116 & & NC\_013991.2 & Phoenix dactylifera \\
117 & & NC\_016068.1 & Nicotiana undulata \\ [1ex]
118 %Entering ninth group
119 & & NC\_009618.1 & Cycas taitungensis \\
120 & & NC\_011942.1 & Gnetum parvifolium \\
121 & & NC\_016058.1 & Larix decidua \\
122 Gymnosperms & 7 & NC\_016063.1 & Cephalotaxus wilsoniana \\
123 & & NC\_016065.1 & Taiwania cryptomerioides \\
124 & & NC\_016069.1 & Picea morrisonicola \\
125 & & NC\_016986.1 & Gingko biloba \\ [1ex]
126 %Entering tenth group
127 Haptophytes & 1 & NC\_007288.1 & Emiliana huxleyi\\ [1ex]
128 %Entering eleventh group
129 Lycophytes & 2 & NC\_014675.1 & Isoetes flaccida \\
130 & & NC\_006861.1 & Huperzia lucidula \\
134 \subsection{Gene Extraction Techniques from annotated NCBI genomes}
135 With NCBI, the idea is to use the existing annotations of NCBI with chloroplast genomes. To extract the core and pan genes: Core extraction techniques with NCBI are based on two techniques: Gene count and Gene contents based on some similarity issues.
137 \subsubsection{Core genes based on NCBI Gene names and Counts}
138 The trivial and simple idea to construct the core genome is based on the extraction of Genes names (as gene presence or absence). For instant, in this stage neither sequence comparison nor new annotation were made, we just want to extract all gene counts stored in each chloroplast genome then find the intersection core genes based on gene names.\\
140 \textbf{Step I: pre-processing}\\
141 The objective from this step is to organize, solve genes duplications, and generate sets of genes for each genome. The input to the system is a list of genomes from NCBI stored as \textit{.fasta} files that include a collection of Protein coding genes\cite{parra2007cegma,RDogma}(genes that produce protein) with its coding sequences.
142 As a preparation step to achieve the set of core genes, we need to translate these genomes using \textit{BioPython} package\cite{chapman2000biopython}, and extracting all information needed to find the core genes. The process starts by converting each genome in fasta format to GenVision\cite{geneVision} format from DNASTAR, and this is not an easy job. The output from this operation is a lists of genes stored in a local database for genomes, their genes names and genes counts. In this stage, we will accumulate some Gene duplications with each genome treated. In other words, duplication in gene name can comes from genes fragments as long as chloroplast DNA sequences. We defines \textit{Identical state} to be the state that each gene present only one time in a genome (i.e Gene has no copy) without considering the position or gene orientation. This state can be reached by filtering the database from redundant gene name. To do this, we have two solutions: first, we made an orthography checking. Orthography checking is used to merge fragments of a gene to form one gene.
143 Second, we convert the list of genes names for each genome (i.e. after orthography check) in the database to be a set of genes names. Mathematically speaking, if $G=\left[g_1,g_2,g_3,g_1,g_3,g_4\right]$ is a list of genes names, by using the definition of a set in mathematics, we will have $set(G)=\{g_1,g_2,g_3,g_4\}$, and $|G|=4$ where $|G|$ is the cardinality number of the set $G$ which represent the number of genes in the set. With NCBI genomes, we do not have a problem of genes fragments because they already treated it, but there are a problem of genes orthography. In our method, this can generate the problem of gene lost and effect in turn the core genes.
144 The whole process of extracting core genome based on genes names and counts among genomes is illustrate in Figure \ref{Fig2}.
148 \includegraphics[width=0.9\textwidth]{NCBI_GeneName}
150 \caption{Extracting Core genes based on NCBI Gene name and Counts}
154 \textbf{Step II: Gene Intersection}\\
155 The goal of this step is trying to find maximum core genes from sets of genes in the database. The idea for finding core genes is to collect in each iteration the maximum number of common genes. To do this, the system build an \textit{Intersection core matrix(ICM)}. ICM here is a two dimensional symmetric matrix where each row and column represent a set of genes for one genome in the local database. Each position in ICM stores the \textit{intersection scores}. Intersection Score(IS), is the cardinality number of a core genes comes from intersecting in each iteration the set of genes for one genome with all other gene sets belong to the rest of genomes in the database. Taking maximum cardinality from each row and then taking the maximum of them will result to select the best two genomes with their maximum core. Mathematically speaking, if we have an $mxn$ matrix where $m,n=$number of genomes in database. lets consider $Z=max_{i<j}(\vert x_i \cap x_j\vert)$ where $x_i, x_j$ are sets of row and column elements in a matrix. if $Z=0$ then we have \textit{disjoint relation} (i.e no common genes between to genomes). In this case the system ignore the set of genes that smash the core genes. Otherwise, The system remove these two genomes from ICM and add new core with a coreID of them to ICM for the calculation in next iteration. The partial core genes generated with its set of genes will store in a database for reused for drawing the tree. this process repeat until all genomes treated.
156 We observe that ICM will result to be very large because of the huge amount of data that it stores. In addition, this will results to be time and memory consuming for calculating the intersection scores by using just genes names. To increase the speed of calculations, we can calculate the upper triangle scores only and exclude diagonal scores that is why we write $max_{i<j}$. This will reduce whole processing time and memory to half. The time complexity for this process after enhancement changed from $O(n^2)$ to $O((n-1)\log{n})$.\\
158 The Algorithm of construction the matrix and extracting maximum core genes where illustrated in Algorithm \ref{Alg1}. The output from this step is the maximum core genes with its genomes to draw it in a tree.
161 \caption{Extract Maximum Intersection Score}
164 \REQUIRE $L \leftarrow sets of genomes genes$
165 \ENSURE $B1 \leftarrow Max core$
166 \FOR{$i \leftarrow 0:len(L)-1$}
167 \STATE $core1 \leftarrow set(GenomeList[L[i]])$
168 \STATE $score1 \leftarrow 0$
169 \STATE $g1,g2 \leftarrow$ " "
170 \FOR{$j \leftarrow i+1:len(L)$}
171 \STATE $core2 \leftarrow set(GenomeList[L[i]])$
173 \STATE $Core \leftarrow core1 \cap core2$
174 \IF{$len(Core) > score1$}
175 \STATE $g1 \leftarrow L[i]$
176 \STATE $g2 \leftarrow L[j]$
177 \STATE $Score \leftarrow len(Core)$
178 \ELSIF{$len(Core) == 0$}
179 \STATE $g1 \leftarrow L[i]$
180 \STATE $g2 \leftarrow L[j]$
181 \STATE $Score \leftarrow -1$
185 \STATE $B1[score1] \leftarrow (g1,g2)$
191 \textit{GenomeList} represents the database.\\
193 \textbf{Step III: Drawing the Tree}\\
194 The goal here is to visualizing the results by build a tree of evolution. The system can produce this tree automatically by using Dot graphs package\cite{gansner2002drawing} from Graphviz library and all information available in a database. Core genes generated with their genes can be very important information in the tree, because they can viewed as an ancestor information for two genomes or more. Further more, each node represents a genome or core as \textit{(Genes count:Family name, Scientific names, Accession number)}, Edges represent numbers of lost genes from genomes-core or core-core relationship. The number of lost genes here can be an important factor for evolution, it represents how much lost of genes for the species in same or different families. By the principle of classification, small number of gene lost among species indicate that those species are related together and belong to same family, while big genes lost means that species is far to be in the same family. To see the picture clearly, Phylogenetic tree is an evolutionary tree generated also by the system. Generating this tree is based on the distances among genes sequences. There are many resources to build such tree (for example: PHYML\cite{guindon2005phyml}, RAxML{\cite{stamatakis2008raxml,stamatakis2005raxml}, BioNJ , and TNT\cite{goloboff2008tnt}}. We consider to use RAxML\cite{stamatakis2008raxml,stamatakis2005raxml} to generate this tree.
196 The main drawback from this method is that we can not depending only on genes names because of three causes: first, the genome may have not totally named (This can be found in early versions of NCBI genomes), so we will have some lost sequences. Second, we may have two genes sharing the same name, while their sequences are different. Third, we need to annotate all the genomes.
198 \subsubsection{Extracting Core genome from NCBI gene contents}
201 \subsection{Core genes from Dogma Annotation tool}
202 In previous section, extracting core genes based on NCBI annotation caused some lost of genes due to annotation process. Annotation can play an important role for these losts, because it represents the first process of gene identification. Good annotation tool still be challenged subject. (Genis Parra in 2007) published a paper state that the subject of accurately genomic and/or gene annotation is still an open source problem, even in the best case scenario where any project has all the expert biologists resources to annotate gene structures, the catalogues of genes can still unclear and still less accurate than experts. Where \cite{Bakke2009} also state ("Errors in the annotations are routinely deposited in databases such as NCBI and used to validate subsequent annotation errors."). So, good core genes still needs good annotation tool. A lot of software today’s were developed for extracted core genes for eukaryote and prokaryote organisms such as CEGMA\cite{parra2007cegma}, Coregenes 3.0\cite{zafar2002coregenes}, and Dogma\cite{RDogma}. The appropriate annotation tool for plant chloroplast and mitochondrial genomes is Dogma.
204 \subsubsection{Why Dogma rather than NCBI annotation?}
205 Dogma is an annotation tool developed in the university of Texas by \cite{RDogma} in 2004. Dogma is an abbreviation of \textit{Dual Organellar GenoMe Annotator}\cite{RDogma} for plant chloroplast and animal mitochondrial genomes.
206 It has its own database for translated the genome in all six reading frames and query the amino acid sequence database using Blast\cite{altschul1990basic}(i.e Blastx) with various parameters, and to identify protein coding genes\cite{parra2007cegma,RDogma} in the input genome based on sequence similarity of genes in Dogma database. Further more, it can produce the \textit{Transfer RNAs (tRNA)}\cite{RDogma}, and the \textit{Ribosomal RNAs (rRNA)}\cite{RDogma} and verifying their start and end positions rather than NCBI annotation tool.
208 \subsubsection{Core genes based on Dogma Genes names and counts}
209 The main goal is to get as much as possible the core genes of maximum coding genes. According to \cite{Bakke2009}, annotation method like dogma can give us more reliable coding genes than NCBI. This is because NCBI annotation can carry some annotation and gene identification errors. The general overview of whole process of extraction illustrated in fiqure \ref{dog:Fig1}, the pipeline of extracting core genes can summarize in the following steps:\\
213 \includegraphics[width=0.7\textwidth]{Dogma_GeneName}
215 \caption{Core genome based on Dogma Gene Name and count}
218 \textbf{Step I: Pre-processing}\\
219 Pre-processing step represents the key difference between methods of extracting core genes. In figure \ref{dog:Fig1}, The pre-processing step can be divided into two tasks: First, It starts to annotate our genome samples in the form of complete genomes (i.e \textit{Unannotated genomes} from NCBI by using Dogma web tool. The whole annotation process by using dogma website is done manually. The output from the annotation process is considered to be a collection of coding genes file for each genome in the form of GeneVision\cite{geneVision} file format.\\
220 Where the second task is to solve gene fragments. Defragment process starts here to solve fragments of coding genes for each genome, this process can avoid gene duplication. All genomes now are fully annotated, their genes were de-fragmented, and genes list and counts identified. These information stored in local database.\\
222 From these two tasks, we can obtain clearly one copy of coding genes. To ensure that genes produces from dogma annotation process is same as the genes in NCBI. We apply separately a quality check process that align the same gene from dogma and NCBI with respect to a specific threshold.\\
224 \textbf{Step II: Extraction Core genes}\\
225 Extracting core genes will use the same process presented in the section of extracting core genes based on NCBI genes. ICM matrix will be considered by calculating the upper triangular cardinality cores to save time and to find the maximum length of core genes from each iteration see algorithm \ref{Alg1}. From each iteration, two genomes are considered to draw with their maximum cardinality core genes until no genome remain in the database. The key point here is that the intersection genome that smash the core genes in each iteration will be ignored from this competition.\\
227 \textbf{Step III: Draw the tree}
228 To build the tree of evolution for genomes. The algorithm is considered to take from the data base the first coreID generated from step two and draw sequentially all the genomes that create this core. Sometimes, we have a core genome that intersect with another one. This tree also represented as a set of nodes which represent genome names and a set of edges, which represent the number of gene lost from each genome. Phylogenetic tree also considered here by using RAxML{\cite{stamatakis2008raxml,stamatakis2005raxml} based on calculating the distances among core genes in the database.
230 \subsubsection{Core genome from Dogma gene contents}
234 \includegraphics[width=0.7\textwidth]{Dogma_GeneContent}
236 \caption{Core genes based on the comparison of Dogma Genes Sequences}
