حضرت خان بابا ؒدی نذر
تیرے در تے میں بابا آئی کدوں دی کھڑی
پاویں جھات کرم دی آوے دید دی گھڑی
تیرے در توں سوالی کدے جان نہ خالی
میں پکڑ کے جالی بابا کدوں دی کھڑی
توں ایں خواجہ اجمیری دا بوہت پیارا
تونسے والیاں دی اکھیاں دا توں ایں ستارا
میری شوہ وچ کشتی ، نہیوں دسدا کنارا
قبلہ عالم دے پیارے جاوے غم دی گھڑی
تیرے در تے میں بابا آئی کدوں دی کھڑی
ہور کج وی نہ منگاں ، بخشو اپنی غلامی
ایس در تے ای گزرے میری عمر تمامی
تیرے جھاڑو میں دیساں ایہو بھردی میں حامی
تیرا تک کے دوارہ ، در تیرے آ وڑی
تیرے در تے میں بابا آئی کدوں دی کھڑی
جنھاں یادِ خدا وچ ، ساری زندگی گزاری
ناں اونہاں دا لیاں ٹلدے دکھ تے بیماری
سیاں کرن سلاماں ایتھے پیاں وارو واری
جمعرات نوں لگے ایتھے رونق بڑی
تیرے در تے میں بابا آئی کدوں دی کھڑی
تساں قادری سائیںؔ تے وی کرم کمایا
بخش اوہنوں امامت، اوہدا شان ودھایا
خاص کیتی عنایت ، درشن چا کرایا
وسے قادریؔ اُتے انجے رحم دی جھڑی
تیرے در تے میں بابا آئی کدوں دی کھڑی
It is with profound pleasure that we write this editorial to welcome you to the new journal, “Pakistan Biomedical Journal” (PBMJ), an interdisciplinary international journal. PBMJ has successfully completed its first volume and now its the second volume. We greatly appreciate the response of scientists who have contributed previously and are still contributing to this new journal. The subject of the journal is interesting and we try to address the health related concerns of public and improve the understandingof scientific phenomenons by researchers. Research discoveries are happening at a fast pace, in all the fields and PBMJ provides an ideal forum for exchange of scientific knowledge in terms of full length papers, surveys, reviews, case studies, letters to editor and systematic analysis. PBMJ is committed to publishing all manuscripts receiving a high recommendation from reviewers. The intention of PBMJ is to create space for generation of new knowledge, debate, collaborations among national and international scientists. Our vision is to promote research that will be helpful for knowledge sharing, new discoveries, development of critical thinking among the upcoming scholars, guidance for policy makers, awareness among the concerned community and ultimately benefitting the general population in improving health and fitness at large. It is a matter of pride for us to haveexcellent editorial board members from renowned institutes. We aim to have the best standards of quality of the published manuscripts. With every issue, we are continuously trying to improve the standards. We look forward for more exciting researches and scientific studies from all over the world. We would like to extend a very warm welcome to the readers of PBMJ and hope you will join us as authors, reviewers and editors in future.
Biological sequences consist of A C G and T in a DNA structure and contain vital information of living organisms. This information is used in many applications such as drug design, microarray analysis and phylogenetic trees. Advances in computing technologies, specifically Next Generation Sequencing technologies have increased genomic data at a rapid rate. The increase in genomic data presents significant research challenges in bioinformatics, such as sequence alignment, short read error correction, phylogenetic inference etc. Various tools and algorithms have been proposed for phylogenetic inference. Early algorithms used sequential programs to solve the problem of phylogenetic inference. Improvements were gained in terms of tree accuracy and execution time, however; the programs were still slow, and improvements were needed to infer correct phylogeny in short times. This challenge introduced parallel and distributed processing to the field of bioinformatics. Many tools and programs have been developed based on parallel and distributed computing. This thesis presents algorithmic solutions for phylogenetic inference. Solutions include ‘PhyloDoop’ and ‘SeqCompress’ algorithms. PhyloDoop algorithm is used for inference of phylogenetic trees. The algorithm is based on Maximum Likelihood method, implemented on Hadoop Map/Reduce framework. PhyloDoop is based on clusters i.e. divides the input alignment to clusters, builds trees for each cluster, merges and optimizes all sub-trees and the final tree is also optimized. PhyloDoop is compared to well-known algorithms both on real and simulated datasets. Experiments on real datasets were performed to test likelihood values, execution time, and speedup in distributed environment. The results show better accuracy as compared to other algorithms on most of the datasets. Execution time is also short on most datasets. The proposed algorithm yields better speed up on large datasets. Simulated datasets were used to measure topological accuracy. PhyloDoop is topologically accurate on most datasets with short execution time in comparison to other algorithms. SeqCompress is used to compress DNA sequences in order to reduce memory requirements and execution time. Impressive results are shown in comparison to other algorithms. These results show a gap for efficient usage of compression techniques to infer correct phylogeny with low memory requirements as well as execution time.