ffmpeg

April15

Loads of pain converting simulation output (ppm and TIF) time series into a movie format that a journal likes has led me to start this page. Heres a list of standard ffmpeg (Mac OSX 10.12.3, Terminal) tools used on Terminal with standard tools. You might need to install some of them if you are on some flavour of Linux (or not).

Combining movies and adding text to them

For a recent paper we had to concatenate few movies and add text to them. These are the commands that worked for me:

  1. Combining 3 movies horizontally:
    ffmpeg -i MovieS2A-sm.mov -i MovieS2B-sm.mov -i MovieS2C-sm.mov -filter_complex hstack=inputs=3 MovieS2ABC.mov
    For vertical replace hstack with vstack.
    Inputs=3 indicates the number of input files
  2. Finding the file size using VLC: SIZE 1440×482 (pixels)
  3. Adding a text legend to them in sequence
    ffmpeg -i MovieS2ABC.mov -vf "drawtext=text='A':x=10:y=10:fontsize=24:fontcolor=black" -c:a copy MovieS2ABC-Atxt.mov
    ffmpeg -i MovieS2ABC-Atxt.mov -vf "drawtext=text='B':x=490:y=10:fontsize=24:fontcolor=black" -c:a copy MovieS2ABC-Btxt.mov
    ffmpeg -i MovieS2ABC-Btxt.mov -vf "drawtext=text='C':x=970:y=10:fontsize=24:fontcolor=black" -c:a copy MovieS2ABC-Ctxt.mov

Fin! Samapt! Submit!

Installing FFMPEG

Caveat: I’m using a MAC OSX 10.12.5 (Sierra) with Xcode and multiple gnu developer tools installed using Mac Ports.

1) Convert .avi files to MAC readable (not just VLC) .mp4 files:

ffmpeg.exe -i %d.png -f mp4 -vcodec libx264 -pix_fmt yuv420p test.mp4

The .png file could be an input .avi file (in which case drop %d.png for your file series)

Convert to MOV using QuickTime->Export.

2) Convert pre-existing movie files in Quicktime .mov format to uncompressed AVI or even MAC compatible .mov

ffmpeg -i terasaki.mov -vcodec rawvideo -y terasaki.avi

I recently needed this to make Supplementary Materials files ImageJ readable and neeed UNCOMPRESSED avi files (all compression codecs seemed to trip up ImageJ). The output looks like this:

ffmpeg version 1.2.1 Copyright (c) 2000-2013 the FFmpeg developers
built on Jun 12 2013 13:46:16 with Apple clang version 4.1 (tags/Apple/clang-421.11.66) (based on LLVM 3.1svn)
configuration: --prefix=/opt/local --enable-swscale --enable-avfilter --enable-libmp3lame --enable-libvorbis --enable-libopus --enable-libtheora --enable-libschroedinger --enable-libopenjpeg --enable-libmodplug --enable-libvpx --enable-libspeex --enable-libass --enable-libbluray --enable-gnutls --enable-libfreetype --mandir=/opt/local/share/man --enable-shared --enable-pthreads --cc=/usr/bin/clang --arch=x86_64 --enable-yasm --enable-gpl --enable-postproc --enable-libx264 --enable-libxvid
libavutil      52. 18.100 / 52. 18.100
libavcodec     54. 92.100 / 54. 92.100
libavformat    54. 63.104 / 54. 63.104
libavdevice    54.  3.103 / 54.  3.103
libavfilter     3. 42.103 /  3. 42.103
libswscale      2.  2.100 /  2.  2.100
libswresample   0. 17.102 /  0. 17.102
libpostproc    52.  2.100 / 52.  2.100
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'terasaki-mk0780641002.mov':
Metadata:
major_brand     : qt
minor_version   : 0
compatible_brands: qt
creation_time   : 2017-06-19 06:11:05
encoder         : Mac OS X v? (AVF 1046.9.12, CM 1731.15.207, x86_64)
encoder-eng     : Mac OS X v? (AVF 1046.9.12, CM 1731.15.207, x86_64)
Duration: 00:00:04.67, start: 0.000000, bitrate: 926 kb/s
Stream #0:0(eng): Video: h264 (High) (avc1 / 0x31637661), yuv420p, 370x220 [SAR 1:1 DAR 37:22], 924 kb/s, 6 fps, 6 tbr, 60 tbn, 120 tbc
Metadata:
creation_time   : 2017-06-19 06:11:05
handler_name    : Core Media Data Handler
Output #0, avi, to 'terasaki-mk0780641002.avi':
Metadata:
major_brand     : qt
minor_version   : 0
compatible_brands: qt
encoder-eng     : Mac OS X v? (AVF 1046.9.12, CM 1731.15.207, x86_64)
ISFT            : Lavf54.63.104
Stream #0:0(eng): Video: rawvideo (I420 / 0x30323449), yuv420p, 370x220 [SAR 1:1 DAR 37:22], q=2-31, 200 kb/s, 6 tbn, 6 tbc
Metadata:
creation_time   : 2017-06-19 06:11:05
handler_name    : Core Media Data Handler
Stream mapping:
Stream #0:0 -> #0:0 (h264 -> rawvideo)
Press [q] to stop, [?] for help
frame=   28 fps=0.0 q=0.0 Lsize=    3345kB time=00:00:04.66 bitrate=5871.8kbits/s
video:3339kB audio:0kB subtitle:0 global headers:0kB muxing overhead 0.188078%

For the mac OSX compatible movie files (quicktime playable), a small conversion with decocer fix:

>>ffmpeg -i videoS2.mov -pix_fmt yuv420p videoS2-v3.mov

Falls under the entry Encodingfordumbplayers


References:

1) Stackoverflow: http://stackoverflow.com/questions/14430593/encoding-a-readable-movie-by-quicktime-using-ffmpeg

2) FFMPEG User-List http://www.ffmpeg-archive.org/Uncompressed-AVI-to-Uncompressed-Quicktime-td941357.html

3) https://trac.ffmpeg.org/wiki/Encode/H.264#Encodingfordumbplayers

February11

A recently organized workshop at Pune university virtually invited me to speak about iGEM and synthetic biology. It was an opportunity for me to do something I’ve been meaning to do for a while- have a dialogue with some of my former students who were part of the team. The video of the dialogue is posted here [pune-univ-synthbio].

Label Free Microtubule Microscopy

January5

The exciting work of Mohammed Mahamdeh and Joe Howard (J.Microsc. https://doi.org/10.1111/jmi.12744) prompted me in the December break from teaching in 2018 to try and reproduce it, using the Nikon TiE inverted epifluorescence microscopy in the lab. With the able presence of a talented Masters student Yash Jawale (now PhD student in the Netherlands), we attempted to uncover its secret levers of light attentuation. After frustrating months, we had a visitor from France, Dr. Kheya Sengupta from CINAM Marseille. As an expert on IRM and RICM (a related method) she helped us achieve it- label free microscopy of the 30 nm wide single MT filaments in (IRM)! And since then we have been toying with it, getting better images with passing days. Now if only the remaining experiments would work!

Volunteering for COVID

June20

The pandemic has affected everything and everybody to different degrees. And academics and research is no exception. It is all the more amazing that in this crisis moment, many of the lab members – Kunalika and Neha- have been volunteering to work at the COVID-Testing Centre. This centre was established under the guidance of Dr. Anjan Banerjee and multiple faculty, to address the need for scaling up virus testing in the city of Pune. If we are to have a fighting chance to open up safely, we need to know who is infected and isolate and treat them, till they get better.

This raises an important question about applied and basic research. In our ordinary lives we are at the forefront of (sometimes) obscure problems, driven by our curiosity and perhaps the interest of a few others in the world. Molecular diagnostics are almost the opposite. The methods are set by the national testing agency. It is important to follow protocol and not deviate and to work in large teams not thinking independently.

So the question is are free thinking fundamental researchers capable of harnessing the training in biology to contribute to the need for skilled work. The answer from the IISER Pune experience is a resounding yes. And its a moment of joy for me to see how many students not just those from my lab showing that they are willing to enter these rooms where the virus is being brought and dealing with it scientifically. Just for this, I think our institute’s ethos and sense of team-work, and responsibility to those around us is reaffirmed.

With more such brave and trained souls, the day is not far when we will surmount this pandemic.

Advice on asking me for recommendation letters

June9

Since I have had to clarify this on numerous occasions to multiple people who’ve passed through my lab, this is a convenient place to put all these thoughts in one place. Also acknowledgement to the legendary John Eisen (of metagenomics fame) and his blog-post on exactly this topic.

1) Please email me your request atleast 1 week before your deadline to have sent it.

2) I need to see an updated CV (or resume) so that my facts about you are accurate.

3) Mention from when to when you have worked with me, if you are no longer in my lab. If you are still in my lab, please still mention your start date. You should add a brief summary what you did in that period (or are doing).

4) Please clarify what is the name/title and nature of the position that you are applying for. Specify the potential role you will play in it, if you were hired/recruited/placed. If a statement of purpose (SOP) is associated with it, please share the gist (summary) of it.

5) Provide all information I will need. For a brief moment, put yourself in my shoes. To be able to send the letter, you need to assist me- i.e.

  • If it’s by email then provide me the address.
  • If it’s as a hard copy then provide me the address of the recepient and whatever title-details are required.
  • Often these letters need to be in certain formats. Please send me this. In short, make my work simpler, so I can help you more effectively.
  • If the agency/organization/university has an online portal, please provide my name and email address ONLY AFTER you have taken my approval.

6) Providing a letter of recommendation is part of my job. It will be a recommendation and an honest assessment.

7) Your work with me or interactions in my class, depending on whether you worked with me or took my courses, will determine the nature of what I write. Think of it as कर्म

8 ) Should you choose to ask for letters from phd-student/technicians/project assistants/postdocs in the lab, please find out from the agency that has requested them what criteria they use for determining the appropriateness of the letter. Typically for academic/research positions a supervisory role is required, but not essential for obtaining recommendations.

9) Please do not add me as a referee without asking me first.

Newly minted PhD: Anushree

June3

With excellent work on automated kymography, and more recent work on label-free tracking in DIC microscopy, Anushree Chaphalkar defended her thesis today and richly earned a commendation from the examiners. Congratulations Dr. Chaphalkar!

AmtraK GUI

Anushree’s work was also covered in the local press in 2016. Look forward to more good news.

Visitor from Rome

May13

Garima Verma, a PhD student candidate at the University of Rome, Italy will be visiting us for a few months. Garima is working in the lab of Dr. Grigioni Mauro, Istituto Superiore di Sanità, e Umberto Nanni del Dipartimento di Ingegneria Informatica.

Immunotherapy: the next frontier in rational anti-cancer drugs?

March30

The recent approvals of chimeric antigen receptor t-cell (CART) therapy for commercial use has triggered renewed interest in both the immune-biology and molecular details of how cancers evade the immune system, but also generally the utility of “Synthetic Cell Biology” in “live therapeutics”.  I like to distinguish “live therapeutics” from those that aren’t alive in terms of the complexity. And given the idea of a cell as the smallest unit of life, this is all the more interesting. The “poison”, “burn” and “kill” strategy of the magic bullet (Paul Ehrlich) has a long and effective history. However, for cancer which is a multifactorial non-communicable disease, this paradigm has perhaps been superseded by an approach based on our understanding of the biology, and our ability to manipulate endogeneous functions. While ImmunoTherapeutics (1) are already showing promising results, my own interest is piqued by our ability to convert a molecular understanding of immune cell molecular biology to modify our own cells (2).

REFEFERENCES:
1) https://www.cancer.gov/publications/dictionaries/cancer-terms/def/immune-checkpoint-inhibitor
2) https://www.cancer.gov/about-cancer/treatment/research/car-t-cells#living-drug

Synthetic morphogenesis at EMBL Heidelberg

March30

The use of synthetic biology in engineering biological systems has been rapidly expanding. the conference at EMBL Heidelberg from 17-20 March 2019 on “Synthetic Morphogenesis: From Gene Circuits to Tissue Architecture” highlighted this in the context of understanding growth and developmental morphogenesis. It brought together an unusual combination of researchers ranging from:

  • in vitro reconstitution of cytoskeletal networks
  • giant unilammelar vesicles (GUVs) for encapsulating proteins: towards synthetic cells
  • rebuilding gradients of morphogens by engineering cells
  • organoid models of tissue morphogenesis
  • engineering blastulas and developing embryos predictively modify developmental outcomes

One, many and the nanometer collective: Dynein gliding assays show the way

January8

Our 5 year long study on dynein can be summarized in the “one, many and collective” phrase. We have been able to show that teams of motors on a surface in a gliding assay appear to transport different lengths of microtubules differenty. In challenging experiments that went beyond qualitative, Kunalika Jain from the lab optimized a quantitative gliding assay that allows her to infer the change in 2D directionality of transpoted MTs.

Setup of the dynein gliding assay in simulation and experiment

Dynein Collective Transport: Simulation & Experiment (Jain et al. 2019 Soft Matter)

Given the details available at a single molecule level, we asked whether counter-intuitive inferences could be made about the role of length-dependent transport. Turns out due to the peculiar detachment mechanics and “search and capture”, MTs are indeed transported in a more directional manner when 3 um long filaments encounter approximately 10 motors or more. Why 10, why 3 um,.. read more in our paper by Jain, Khetan and Athale (2019) Soft Matter.

« Older Entries