Md. Khaledur Rahman

CRISPRpred: A flexible and efficient tool for sgRNAs on-target activity prediction in CRISPR/Cas9 systems

The CRISPR/Cas9-sgRNA system has recently become a popular tool for genome editing and a very hot topic in the field of medical research. In this system, Cas9 protein is directed to a desired location for gene engineering and cleaves target DNA sequence which is complementary to a 20-nucleotide guide sequence found within the sgRNA. A lot of experimental efforts, ranging from in vivo selection to in silico modeling, have been made for efficient designing of sgRNAs in CRISPR/Cas9 system. In this article, we present a novel tool, called CRISPRpred, for efficient in silico prediction of sgRNAs on-target activity which is based on the applications of Support Vector Machine (SVM) model. To conduct experiments, we have used a benchmark dataset of 17 genes and 5310 guide sequences where there are only 20% true values. CRISPRpred achieves Area Under Receiver Operating Characteristics Curve (AUROC-Curve), Area Under Precision Recall Curve (AUPR-Curve) and maximum Matthews Correlation Coefficient (MCC) as 0.85, 0.56 and 0.48, respectively. Our tool shows approximately 5% improvement in AUPR-Curve and after analyzing all evaluation metrics, we find that CRISPRpred is better than the current state-of-the-art. CRISPRpred is enough flexible to extract relevant features and use them in a learning algorithm. The source code of our entire software with relevant dataset can be found in the following link: https://github.com/khaled-buet/CRISPRpred.

Prefix and suffix transreversals on binary and ternary strings

The problem of sorting by a genome rearrangement event asks for the minimum number of that event required to sort the elements of a given permutation. In this paper, we study a variant of the rearrangement event called prefix and suffix transreversal. A transreversal is an operation which reverses the first block before exchanging two adjacent blocks in a permutation. A prefix (suffix) transreversal always reverses and moves a prefix (suffix) of the permutation to another location. Interestingly, we will apply transreversal not on permutations but on strings over an alphabet of fixed size. We determine the minimum number of prefix and suffix transreversals required to sort the binary and ternary strings, with polynomial time algorithms for these sorting problems.

Finding suitable places for live campaigns using location-based services

In the recent years, the idea of reaching customers through human experience has triggered a new marketing strategy known as live campaigns. We can expect that live campaigns will become more pervasive and profitable, but not before addressing key business challenges. It can be easily ruined if the campaign agency fails to identify the optimal location and time. In this paper, we address the challenge of finding a suitable location from online location based services for arranging a live campaign according to given schedule among a set of candidate locations. We study the predictive power of various spatio-temporal mining features on the capability of gathering audience through the use of a dataset collected from Foursquare of New York City. Finally, we develop models which will predict the expected audience at a location based on these features. We achieve 50.46% accuracy in individual feature based approach and an accuracy of 72.6% in Support Vector Machine (SVM) regression model.

Pancake flipping and sorting permutations

In this paper, we study several variations of the pancake flipping problem, which is also well known as the problem of sorting by prefix reversals. We consider the variations in the sorting process by adding with prefix reversals other similar operations such as prefix transpositions and prefix transreversals. We first study the problem of sorting unsigned permutations by prefix reversals and prefix transpositions and present a 3-approximation algorithm for this problem. Then we give a 2-approximation algorithm for sorting by prefix reversals and prefix transreversals. We also provide a 3-approximation algorithm for sorting by prefix reversals and prefix transpositions where the operations are always applied at the unsorted suffix of the permutation. We further analyze the problem from practical point of view and show quantitatively how approximation ratios of our algorithms improve with the increase of number of prefix reversals applied by optimal algorithms. Finally, we present experimental results to support our analysis.

Sponge: A searchable P2P mobile app store using DHTs

In this paper we propose a novel searchable P2P content storage for storing mobile apps. In the growing need of storing and distributing huge number mobile apps across large number of users, it has been deemed necessary to think of a storage that leverages enormous amount of content space that common people can share these days. In our proposal, we leverage popular P2P stores, such as standard DHTs, for storing apps. One problem with DTHs is that they support flat names (denoted as keys) for content objects, which do not make content objects searchable by their different attributes. When users look for a certain app in an app store, they do not necessarily look by their unique names, instead by a set of attribute-value pairs (multi-attribute queries). We devise a technique, called predicate hashing, by which we hash these attribute-value pairs into hashed keys that are in turn used to locate corresponding apps from the underlying DHT. By doing this we effectively instrument flat named DHTs into reverse ifndexable content store. To this end, we build Sponge, a Kademlia based P2P app store, and do simulation based experiments to show that Sponge can retrieve content against multi-attribute queries.