Research / statistical analyst post

Skilled programmer and research analyst sought to work on this (the EDCTP funded project “Biostatistical Methods for Longitudinal Analysis of Burden of Lung Health and Tuberculosis in Africa”.) project. You will work with a team of statisticians and analysts to develop simulation and analytic models for complex longitudinal data.

Depending on skills and background plus interests this post could look quite different for different candidates. Please get in touch if you would like to discuss.

Requirements:
• Completed MPH (Epidemiology & Biostatistics specialisation) or MSc in Statistics / Biostatistics
• Strong R programming skills
• Experience with clinical/health data analysis and reporting
Desirable:
• Evidence of scientific writing skills
• PhD in Statistics, Biostatistics or related field
• Bayesian modelling experience

Start date: Immediately
Duration: 9 months (end Nov 30, 2022), with possibility of a 12 month renewal
Payclass: UCT Payclass 8/9 depending on qualifications, prefer full time, but willing to consider part time for right candidate
Closing date: this position will remain open until filled
Note you must be legally permitted to take up employment in South Africa for this post.

This project is part of the EDCTP2 programme supported by the European Union (TMA2017SF-1959).
To apply: please email a single pdf file with your cover letter, CV to maia.lesosky@uct.ac.za

Clustering of longitudinal viral loads in the Western Cape

The abstract and a few summary figures of the MPH mini-dissertation submitted by Eke Arua for his degree.

Abstract

Introduction: Data from routine viral load monitoring is important for assessing the programmatic effectiveness of antiretroviral treatment (ART) in South Africa, but this is usually analysed in a cross-sectional manner and there are few analyses of longitudinal viral load trajectories over time. Cluster analysis was used to identify latent viral load trajectory patterns in a large dataset of routinely collected viral load measurements in the Western Cape, South Africa.

Methods: We analysed available VL measurements collected during routine care from the Western Cape public sector antiretroviral treatment programs, including all people living with HIV who were enrolled on ART and had experienced at least one elevated (>1000 copies/mL) viral load test result between 2008 and 2018. Empirical rules-based classification, nonparametric clustering using the KML-Shape algorithm and model-based clustering using latent class mixture modelling were used to cluster viral load trajectories.

Results: Both the nonparametric and model-based clustering techniques identified five latent viral load trajectory subgroups. The shapes and magnitudes of these subgroups differed according to method. Majority of individuals’ trajectories belonged to clusters that had a decreasing VL trend. Most of the trajectory subgroups identified had prolonged periods of low-level viremia. Both methods identified viral load trajectory clusters that increased over time.

Conclusion: Cluster analysis is a useful tool for identifying latent VL trajectory clusters in the Western Cape and the dynamic VL cluster patterns that exist among the large VL dataset can offer important insights. Further research is needed to understand factors associated with belonging to these clusters to improve population viral load suppression rates and aid HIV prevention.

Figure 1: Sample of 500 individual trajectories with LOESS smoother for different patterns over time: A) single elevated VL not at first visit; B) multiple elevated VL; C) single elevated VL at first visit and D) elevated VL throughout

Figure 2: The non-parametric approach “KLM-shape” identified five VL trajectories, mean trajectory plotted below with the percentage of individuals classified in that type indicated in top right corner.

Figure 3: The trajectories identified by the model based LCMM approach using fractional polynomials.

There are many more tables and details in the mini-dissertation itself, but it was a large and difficult analysis so well done to Eke!

Data planning

This has been up in various guises before, it remains incomplete but I’ve had requests recently to post.

Use this document in conjunction with the Research data planning checklist and the research data guidelines above to help with the research data planning process.

1. Describing your data
Consider:
– How will this research be generated and used in this project?
– Describe the data set as completely as possible. Include information about the format, average size, volume and/or estimated number of data files produced.
– Consider life cycles of this data set:
– What stages does the data go through (eg raw, processed, analyzed)?
– What methodologies at each stage?
– What tools and instruments are used?
– Who is involved (eg. Professors, lab techs, students)
– How will this data be managed?
– How will you identify and cover costs of managing data sets?

 Data collection
– Are you collecting on paper based forms? How will they be digitised?
– Any additional requirements (eg. image scanner, optical character recognition)?
– Collecting data by mobile phone or tablet? What software, how will you access?
– Who is responsible? Where will the data be hosted?

Data and metadata standards
– Are there any standard formats in for eld for managing or disseminating the data sets (eg. XML, ASCII, CSV)?
– Is your format proprietary rather than open and is this essential?
– If there is not a standard format, how will you format the data so that others in your field will be able to make use of it?
– Who in your team will have responsibility for ensuring that data standards are properly applied and data are properly formatted?

Metadata is structured information that describes or otherwise makes it easier to retrieve, use or manage an information resource. It represents the who, what, why, where and how of the resource.
– How will metadata be generated and captures for each of your data sets?
– Are you aware of any metadata standards that could be used for you data sets?
– If there is not a metadata standard, what metadata will you need to generate so that others in your field will be able to find, understand and make use of your data?
– Who in your research team will be responsible for ensuring metadata standards are followed?

2. Intellectual property, ownership and copyright
Funding agencies may have varying approaches toward IP, copyright and related issues. UCT has specific IP and ownership, and South African IP law is different than many other countries.
– Who will own these data sets? Any other stakeholders need to be consulted before data sets are made
available?
– Will you permit re-use of data, either with or without conditions?
– Will you permit re-distribution of the data, either with or without conditions?
– Will you permit the creation and publication of derivatives of the data, either with or without conditions?
– Will you permit others to use the data to develop commercial products or in ways that produce a financial benefit for themselves, either with or without conditions?


3. Data sharing
Funding agencies may recommend or require data sharing during the course of research.
– How will the people who generated the data sets receive attribution for their work?
– Who would be the target audience for your data sets, and how would they use your data?
– When will you share each of your data sets (eg. After data has been normalised, corrected, after publication, etc)?
– Will you place any conditions on the sharing of your data with others (ie requiring some form of acknowledgement or attribution, forbidding for-profit use)?
– If these data sets contain sensitive information, what steps will you take to ensure protection?
– Do you need to get specific consent for data sharing? Note this if often essential for qualitative interviews, genetic data.

Data archiving and preservation
– Which of you data sets have long-term value to others?
– How will you ensure ongoing access beyond the life of the project?
– What related information needs to be preserved with the data?
– How will you or the repository you are working with ensure that these data sets are able to withstand changes in or the obsolescence of the storage techniques?

Additional resources
This document borrows heavily from: Purdue University Libraries. Data management plan self assessment questionnaire. Purdue University, West Lafayette IN. 2/4/11.

Positions available (closing end of Jan)

We have a few positions available

  1. Modeller – we need a great R programmer to work on statistical simulation. This can either be as a postdoc (tax free) or we can appoint as a research fellow (not tax free), either way post PhD as research experience will help. This is really contract programming, but a fun project with lots of scope. The description pasted here is for the ‘postdoc’ post, but feel free to drop an email to discuss the conditions for the other variant. (job description pdf file)
  2. Statistical analyst – MSc level analyst for Bayesian multilevel modelling related to respiratory disease. (job description pdf file)

See job descriptions below.

Job 1: Biostatistics Group – Postdoctoral fellowship in Biostatistics

Specializing in Statistical Simulation

The Division of Epidemiology & Biostatistics at UCT is seeking a Postdoctoral Fellow in Biostatistics to conduct research developing individual stochastic simulation models relating to HIV, ART, and viral load. The successful candidate will have skills in programming and biostatistics, with specific experience using R software to build simulation models. Experience with Rcpp and/or C++ as well as the pacakge ‘shiny’ will be beneficial. He/she will join the Division of Epidemiology & Biostatistics (www.http://www.publichealth.uct.ac.za/phfm_epidemiology-and-biostatistics), which forms the centre of biostatistics activities in the Faculty.

Academic Conditions of Award:

The successful candidate must be in possession of a PhD or equivalent degree in a relevant area (Computer/Data Science, Statistics, Biostatistics), with exposure to biostatistics. The candidate must provide evidence of excellent R programming skills.

The research to be undertaken will include:

  • extension of an existing simulation model to incorporate new data including effects of DTG based ART regimens, and downstream statistical analysis and data interpretation;
  • optimising simulation model by re-coding and/or incorporating Rcpp/C++ components;
  • parallelisation of the simulation model;
  • if interested, the incumbent will have opportunities to lead author manuscripts and other outputs

General Conditions of Award:

(i) Only individuals who have achieved the doctoral degree within the past 5 years are eligible to apply

(ii) Applicants may not previously (since their PhD) have held full-time professional or academic positions

(iii) The successful candidate may, as part of their professional development, be required to participate in departmental activities such as limited teaching and supervision and duties incidental thereto

(iv) The successful incumbent will be required to comply with the University’s approved policies, procedures and practises for the postdoctoral sector.

Value and tenure:

The value of the fellowship is ZAR 300 000 per annum (non-taxed). No benefits or travel allowances are included in the value of the fellowship.

Application process:

Applications and inquiries should be submitted to A/Prof Maia Lesosky (maia.lesosky@uct.ac.za). To apply, please e-mail a letter of application stating areas of expertise, research interests, and experience, a CV including a list of publications, code repositories, copies of academic transcripts, and names (and contact details) of at least two academics who have taught, supervised or worked alongside the applicant.

Only shortlisted candidates will be contacted. A competency evaluation may be applied.

Closing date for applications: 31 Jan 2020

The University of Cape Town reserves the right to disqualify ineligible, incomplete and/or inappropriate applications. The University of Cape Town reserves the right to change the conditions of award or to make no awards at all.

Job 2: Biostatistics Group – Statistical Analyst

The BiostatsLAB-AFRICA group in the Division of Epidemiology & Biostatistics at UCT is seeking a statistical analyst for a 12-month contract post (renewable) to carry out research related work in longitudinal models relating to respiratory disease in sub-Saharan Africa. The successful candidate will have skills in programming and biostatistics, with specific experience using R software and applied data analysis. He/she will join the Division of Epidemiology & Biostatistics (www.http://www.publichealth.uct.ac.za/phfm_epidemiology-and-biostatistics), which forms the centre of biostatistics activities in the Faculty.

Requirements include:

  • Masters degree in statistics, epidemiology or public health (statistics or epidemiology specialisation)
  • Minimum 2 years research and statistical analysis experience in a research setting
  • Highly proficient in R
  • Experience working in a collaborative research environment advantageous
  • Desire to further training through a PhD (and eligibility in that regard) advantageous

Responsibilities include:

  • Carrying out statistical analysis, including development of analysis plans and documentation of analysis results
  • Carrying out statistical modeling, including documentation of modeling
  • Supporting investigators and students with data/statistical analysis
  • Completing statistical analyses as requested
  • Carrying out project management tasks (10% time) as requested
  • Participation in analysis and writing of manuscripts for publication

The annual remuneration package will be based on experience and qualifications, for example:

UCT PC 8/9 (http://www.hr.uct.ac.za/hr/benefits/remuneration/coe_ranges/payclass1_12)

Application process:

Applications and inquiries should be submitted to A/Prof Maia Lesosky (maia.lesosky@uct.ac.za). To apply, please e-mail a letter of application stating areas of expertise, research interests, and experience, a CV including a list of publications, code repositories, copies of academic transcripts, and names (and contact details) of at least two academics and/or supervisors who have taught, supervised or worked alongside the applicant.

Only shortlisted candidates will be contacted. A competency evaluation may be applied.

Closing date for applications: 31 Jan 2020

The University of Cape Town reserves the right to disqualify ineligible, incomplete and/or inappropriate applications. The University of Cape Town reserves the right to change the conditions of appointment or to make no appointment at all.

UCT is committed to the pursuit of excellence, diversity and redress in achieving its Equity Targets.

Our Employment Equity Policy is available at http://www.uct.ac.za/downloads/uct.ac.za/about/policies/eepolicy.pdf

For this post we seek to particularly attract black (i.e. African, Coloured or Indian) South African candidates.


PATS MECOR 2020 – Cameroon

Join us in Douala, Cameroon 20-24 April, 2020 for the annual PATS MECOR course. BiostatsLAB-Africa will again be supporting this course with biostatistics teaching & consulting, but we would also like to offer two workshops to local statisticians. Please register your interest here and we will get in touch.

Workshop 1 Longitudinal modelling with R: A one day interactive workshop developing skills in longitudinal modelling with R.

Workshop 2 (interest permitting): Statistics for science journalists: how to write about research. We will do a half day workshop aimed at science journalists who would like some training or a refresher on how to understand and interpret the numbers coming out of research publications.

If you are an early career respiratory researcher who is interested in the PATS MECOR program, please see details here – applications close 06 January 2020.

Workshops supported by the Academy of Medical Sciences Newton Advanced Fellowship and the European and Developing Countries Clinical Trials Partnership (EDCTP2) programme supported by the European Union (TMA2017SF-1959)

ZivaHub & uploading of some misc work

UCT has a branded version of figshare called ZivaHub and as promised, we are starting to upload some old slide decks, posters, etc. There are a few documents pending review, but eventually you will be able to find under my name here . We have also just uploaded a pre-print onto the arXiv titled “Bias in the estimation of cumulative viremia in cohort studies of HIV-infected individuals”.

Collaborative Biostatistics Training Workshop

We will be running our first workshop focused on early career biostatisticians in June 2019 in Tanzania. Dates are 13-14 June at the White Sands Hotel, Jangwani Beach, Dar Es Salaam. We are currently taking applications.

Topics:

  • The many roles of the collaborating biostatistician
  • Collaboration with clinical researchers (theory & practice)
  • Development of collaboration skills
  • In depth skills assessment
  • Methods & practice: Sample size and power (G*Power software)
  • Biostatistical methods (GLMs, GAMs/splines, GEE and clustered data)
    (may change depending on skills assessment)

Dates: Thursday 13 – Friday 14, June, 2019, 8:30-17:00 daily
Venue: White Sands Hotel, Jangwani Beach, Dar Es Salaam
Cost: $80 (2 days) includes lunch and tea breaks [Note: Does NOT include accommodation or travel]
Apply: https://goo.gl/forms/NxUu9aunt6vo2hSC2 DEADLINE: 1 May, 2019
Decision & feedback will be made on applications by 5 June, 2019

“Introduction to R” Workshops

The first of three “Introduction to R” workshops led by the Division of Epidemiology and Biostatistics in the School started in the second week of March. Over three days approximately 25 Masters and PhD students learned basics of R programming and an introduction to data manipulation in R. The workshop was aimed at new users and started from the very basics.

Future workshops will build on this one and introduce students to more advanced data analysis skills in R. The workshop materials are available through the public (to UCT community) Vula site “R Training (Div Epi Bios)” which can be joined without invitation. Training was led by Elton Mukonda, a PhD student in the Division of Epidemiology and Biostatistics and supported by a number of current 2nd year MPH students as teaching assistants.

Re-post from here

Academy of Medical Sciences Newton Advanced Fellowship awarded to A/Prof Maia Lesosky

Associate Professor Maia Lesosky of the Division of Epidemiology & Biostatistics has been awarded a Newton Advanced Fellowship from the Royal Society, in partnership with the Academy of Medical Sciences for the period 2019-2021.

The Fellowship aims to enable established international researchers with an opportunity to develop the research strengths and capabilities of their research groups through training, collaboration and reciprocal visits with a partner in the UK.

During the Fellowship, Dr Lesosky will collaborate with the Liverpool School of Tropical Medicine (LSTM) and the Centre for Health Informatics, Computing and Statistics (CHICAS) at the University of Lancaster Medical School. The research groups will join forces to extend Bayesian statistical methods for the analysis of multilevel models relating to exposure and lung health. 

Repost from here