Interview with Waldeyr Silva about the 2Path Core Project (Terpenoid Metabolic Network)

London, October 2, 2017

Today Waldeyr Silva accepted to release a quick interview about his project and why he has chosen OrientDB.

Luca: Hi Waldeyr, could you tell us a little about your project?

Waldeyr: The 2path core is a terpenoid metabolic network modeled using a graph database (in this version: OrientDB). 2path also provides a Python program to reconstruct the terpenoid network of a plant organism of interest from genomic data. You can either use the database and explore the terpenoid metabolic network or reconstruct the terpenoid metabolic network of a target organism from protein or nucleotide sequences.

This is a screenshot about the generation of the model (built in python):


And after a while this is the generated database in OrientDB:


Luca: Did you evaluate other DBMS for this project?

Waldeyr: Yes, the first version of 2path was published using Neo4J. In all cases the performance and compatibility with the modeling of the graph we popped up were adequate. However, OrientDB has brought together two essential, mutually exclusive features in the other databases: favorable license and easy portability. The OrientDB license is fully compatible with the free distribution of knowledge. Also, the database can be distributed embedded in some application.

Luca: Why did you choose OrientDB?

Waldeyr: Primarily by license and portability, but also, it is very intuitive to implement the data model with object-oriented features as abstract classes.

Luca: What API are you using with OrientDB?

Waldeyr: The software for the reconstruction of metabolic networks was written in Python. And I used the pyorient library. The software is still under construction, with new features being released soon, but you can do the basics: rebuilding a metabolic network from cDNA or protein sequences.

Luca: In your opinion, which are the best 3 OrientDB features?

Waldeyr: Object-oriented features, License and Portability


Thanks Waldeyr to have shared your experience with the OrientDB Community. If you want to know more about 2Path look at If you’re reading this post and you also would like to share your experience with OrientDB, please drop an email to info -at-


Luca Garulli
VP OrientDB


This week we interviewed Odysseas Pentakalos.


Luca: “Hi Odysseas, Can you tell us a little about your company and your role?”

Odysseas: “Hi Luca. I am the Chief Technology Officer at SYSNET International which is based in Northern Virginia in the United States and focuses primarily on offering consulting services relating to software architecture and large distributed system implementation. For the past seven years we have also focused on the Health Information Technology and we have developed an open source product called OpenEMPI that offers master data management for patient demographic and other data.”

Luca: “What’s your background with other databases?”

Odysseas: “Through our consulting practice we have worked with pretty much every commercial and open source relational database in the market including Oracle, Microsoft SQL Server, MySQL and Postgres. We also used MongoDB on a recent research project for storing clinical data. Earlier versions of OpenEMPI used solely Postgres to persist the data managed by the system but we’ve been working on a different edition of the software that uses OrientDB to store the key data.”

Luca: “Why did you chose OrientDB?”

Odysseas: “In the entity edition of OpenEMPI we wanted for the system to be able to manage records of any type of entity and not just patient demographic data. Before we selected OrientDB we developed a few prototypes to evaluate the performance of the database in being able to efficiently handle large volumes of records of data where the precise field composition of each record is specified by the user during deployment. We evaluated both relational databases and a couple of NoSQL database and OrientDB was clearly the best option. The flexible Apache license was also another key reason why we made the selection. Finally, the fact that OrientDB is supported by such a dedicated group of people who also provide commercial support if needed, was the feature that closed the deal.”

Luca: “What API are you using with OrientDB?”

Odysseas: “We have been working with OrientDB for over a year now. In the early stages we used the OrientDB Java-based Graph API, but earlier this year we migrated to the Blueprints API since the legacy one was no longer supported. We tried to use the Blueprints API as much as possible since it is a standard API for graph databases but in some areas we found that it is lacking the capabilities we need so we dig deeper and use the OrientDB-specific extensions.”

Luca: “In your opinion, which are the best 3 OrientDB features?”

Odysseas: “There are certainly more than 3 great features in OrientDB but here are the top 3.

    1. Performance: if OrientDB was not able to meet the performance requirements of our application the other features wouldn’t matter much so I have to put performance at the top. We have been impressed by the performance of the database in being able to support the key transactions that our software needs to process. We also plan to exploit the clustering capabilities soon which will allows us to achieve higher overall performance.
    2. Flexible Schema: The flexibility of being able to choose whether to use a schema or not is very important. We are exploiting OrientDBs ability to define classes to capture some of the common attributes of our data model while allowing the users to dynamically extend the schema over time.
    3. Transaction support: Some NoSQL databases provide unlimited scalability of the system but trade-off support for ACID transactions. Our application must support ACID semantics so we are willing to trade-off some performance in exchange for always having consistent data.

Congratulations on the great job you guys have done so far with OrientDB and the best of luck in the future.”

Luca: “Thanks Odysseas!”


Read more about Testimonials and Active Customers.

Luca Garulli
CEO at Orient Technologies LTD
the Company behind OrientDB  

This is the first of many interviews with users regarding their usage of OrientDB. We asked to Kashif Masood, Technical Executive at Awal Technologies (, how they used OrientDB. Read more about Testimonials and Active Customers.


Luca: “Hi Kashif, Can you tell us a little about your company and your role?”

Kashif: “Thanks Luca, I am Technical Executive at Awal Technologies based in Islamabad, Pakistan. We provide businesses smart software solutions to businesses globally. We specialize in PHP (Drupal, WordPress, Yii), .Net and Java technologies and Mobile Apps development. We also have products named Awal ERP, Awal HRM, Awal E-Commerce Solution and Awal Educational Institute Mgmt. System. We use Agile methodologies, mostly SCRUM and our own developed Iterative Development Methodology for software development.”

Luca: “What’s your background with other databases?”

Kashif: “We regularly use MySQL for our client projects. We have used Cloudant (CouchDB), HyperTable, HBase, and OrientDB for specialized needs. We fully advocate NoSQL movement, and consider Document and Graph DBs as the most innovative solution invented in the history of data storage.”

Luca: “Why did you chose OrientDB?”

Kashif: “We chose OrientDB because of its Graph data model, its intuitive SQL like language, Apache License, if you don’t want graph then use Document model (schema less), no costly joins like in RDBMSes and future potential. The product which we had to develop required two key concepts: store individual information in the form of documents (which OrientDB provides in a very convenient manner) and relationships between these documents in the form of a graph structure, since we had some social networking type of features, and OrientDB provided the graph model representation and manipulation in a very powerful manner.

I am glad we didn’t have a learn some aliens language like Cypher to traverse and find patterns out of the relationships. It has a simple SQL like language to do the stuff we needed. This same was very difficult and painful to do using BigTable like DBMSs e.g. HyperTable and HBase. Obviously we are not talking about hundreds of petabytes of data here. While MongoDB was notorious about data lose, because of its late data flushing out to disk. So we tried CouchDB, but we very disappointed to see how we had to query data using its improperly implemented Map-Reduce querying feature. Neo4J is a good product to drain out your money and still keep you unsatisfied, due to its non-distributed nature. So we were left with OrientDB as the single and the best choice for our problems solution.”

Luca: “What API are you using with OrientDB?”

Kashif: “We mostly have used OrientDBs Graph APIs, but also document to store and retrieve individual documents.

Luca: “In your opinion, which are the best 3 OrientDB features?”

Kashif: “Naming 3 best features is a very difficult question to answer. Because for us it is difficult to ignore all the features which are best to name just 3 :). But in a sorted list of all its best features, the first 3 features will be:

  1. Graph-Document data models with schema-less nature
  2. SQL like query language
  3. Apache License”

Luca: “Thanks for the interview, Kashif!”


Read more about Testimonials and Active Customers.

Luca Garulli
CEO at Orient Technologies LTD
the Company behind OrientDB  

Unlock the full potential of your enterprise’s data